JP7658395B2 - Air conditioners - Google Patents

Description

The present invention relates to an air conditioner.

For example, an air conditioner has been proposed that is equipped with a human detection sensor that detects the presence or absence of people in the air-conditioned space, and that uses the detection results of the human detection sensor to set a timer for the start time of air-conditioning operation (for example, Patent Document 1). In Patent Document 1, the detection results of the human detection sensor over several days are compared with "rules for determining the presence or absence of a user" to predict the time when a user will change from being absent to being present in the air-conditioned space, and the timer is set to start the air-conditioning operation based on this prediction result.

Japanese Patent Application Publication No. 11-14121

Meanwhile, in recent years, air conditioners are increasingly being equipped with multi-functions that, in addition to so-called air conditioning operations such as cooling and heating, also have special operations (so-called additional function operations) to keep the air conditioner itself clean. However, simply predicting the time when a user will change from being absent to being present in an air-conditioned space does not allow various operations such as air conditioning and additional function operations to be performed at appropriate times in accordance with the user's daily rhythm (for example, the time to wake up, leave the house, return home, and go to bed).

In view of these problems, the present invention aims to provide an air conditioner that can appropriately perform air conditioning operation and additional function operation (hereinafter referred to as operation) in accordance with the user's lifestyle.

An air conditioner of one embodiment has a sensor, a memory unit, and a behavior prediction unit. The sensor detects the presence or absence of a person in the air-conditioned space. The memory unit uses the detection result of the sensor to associate and store a presence/absence prediction result that predicts the presence or absence of a user in the air-conditioned space with indoor use information that identifies the indoor use of the air-conditioned space. The behavior prediction unit predicts the behavior of the user based on the presence/absence prediction result, the indoor use information of the air-conditioned space, and the time period when air-conditioning operation is expected to start in the air-conditioned space or the time period when air-conditioning operation is expected to stop in the air-conditioned space.

One aspect is that by using the presence/absence prediction results and the indoor use information of the air-conditioned space, it is possible to appropriately operate the air-conditioned space in accordance with the user's lifestyle rhythm.

FIG. 1 is an explanatory diagram illustrating an example of an air conditioning system according to a first embodiment. FIG. 2 is a block diagram showing an example of the configuration of an air conditioner. FIG. 3 is a block diagram showing an example of the configuration of a communications adapter. FIG. 4 is an explanatory diagram showing an example of a prediction result of presence or absence. FIG. 5 is a block diagram showing an example of the configuration of a terminal device. FIG. 6 is a block diagram showing an example of the configuration of a server device. FIG. 7 is an explanatory diagram showing an example of data used to generate a presence/absence pattern. FIG. 8 is an explanatory diagram showing an example of a user pattern predicted using a presence/absence pattern. FIG. 9A is an explanatory diagram showing an example of a timer setting screen of a terminal device. FIG. 9B is an explanatory diagram showing an example of a recommendation screen of a terminal device. FIG. 9C is an explanatory diagram showing an example of a recommended time change screen of a terminal device. FIG. 10 is an explanatory diagram showing an example of a timing chart of each process during the thermal sterilization operation. FIG. 11 is an explanatory diagram showing an example of a timing chart of each process during the filter cleaning operation. FIG. 12A is a flowchart showing an example of a processing operation of the CPU of the server device related to a generation process for generating a presence/absence pattern. FIG. 12B is a flowchart showing an example of the processing operation of the CPU of the server device related to the update processing for updating the presence/absence pattern. FIG. 13A is a flowchart showing an example of processing operations of a CPU of a communications adapter related to recommendation processing. FIG. 13B is a flowchart showing an example of the processing operation of the control unit of the indoor unit related to the air conditioning operation recommendation processing in the living room. FIG. 13C is a flowchart showing an example of the processing operation of the control unit of the indoor unit related to the air conditioning operation recommendation processing in the living room. FIG. 13D is a flowchart showing an example of the processing operation of the control unit of the indoor unit related to the air conditioning operation recommendation process in the bedroom. FIG. 14 is a block diagram illustrating an example of the configuration of an air conditioner according to the second embodiment.

Below, examples of the air conditioner and air conditioning system disclosed in the present application will be described in detail with reference to the drawings. Note that the disclosed technology is not limited to these examples. Furthermore, each of the examples shown below may be modified as appropriate within the scope of not causing any inconsistency.

<Air conditioning system configuration>
Fig. 1 is an explanatory diagram showing an example of an air conditioning system 1 according to Example 1. The air conditioning system 1 shown in Fig. 1 includes an air conditioner 2, a communication adapter 3, a router 4, a server device 5, a relay device 6, a terminal device 7, and a communication network 8.

<Air conditioner configuration>
FIG. 2 is a block diagram showing an example of the configuration of the air conditioner 2. The air conditioner 2 shown in FIG. 2 has an indoor unit 21, an outdoor unit 22, and a remote control 23. The indoor unit 21 is, for example, arranged indoors and is a part of the air conditioner 2 that heats or cools the air in the indoor air-conditioned space. The indoor unit 21 is provided for each air-conditioned space, such as a living room or a bedroom. The indoor unit 21 has a main body 21A, a sensor 21B, a light receiving unit 21C, a control unit 21D, and a memory 21E. The main body 21A is provided with an indoor fan and an indoor heat exchanger (not shown), and the indoor air that has exchanged heat with the refrigerant supplied from the outdoor unit 22 in the indoor heat exchanger is blown out by the indoor fan to heat, cool, dehumidify, etc. the room. The sensor 21B detects the presence or absence of a person in the air-conditioned space. The sensor 21B is, for example, a pyroelectric sensor using infrared rays. When the air conditioner 2 is installed and connected to a commercial power source to supply power, the sensor 21B starts detecting the presence or absence of a person. After this, unless the power supply is stopped, the sensor 21B continues to detect the presence or absence of a person in the air-conditioned space regardless of whether the air conditioner 2 is operating or stopped. The light receiving unit 21C receives a command signal from the remote control 23 and transmits the received command signal to the control unit 21D. The memory 21E is an area for storing various information such as indoor use. The indoor use is information that identifies the use of the air-conditioned space, such as a living room or a bedroom. The indoor use is information used when predicting the user's behavior. The control unit 21D controls the entire indoor unit 21. The control unit 21D executes various commands based on the command signal. The outdoor unit 22 is equipped with, for example, an outdoor fan, a compressor, etc. The remote control 23 is a remote control unit that remotely controls the indoor unit 21 in response to a user's operation.

The control unit 21D has a behavior prediction unit 21D2, a recommendation unit 21D0, and an extraction unit 21D1. The behavior prediction unit 21D2 predicts the user's behavior based on the presence/absence prediction result that predicts the presence/absence of the user in the air-conditioned space and the indoor use information that identifies the indoor use of the air-conditioned space. The presence/absence prediction result is information that accumulates the prediction result of the presence/absence of the user in the air-conditioned space every 10 minutes for 24 hours, which is obtained from the presence/absence prediction unit 34E in the communication adapter 3 described later.

The recommendation unit 21D0 selects an operation to be recommended to the user based on the behavior prediction result of the behavior prediction unit 21D2. The operations include normal air conditioning operations such as cooling operation and heating operation, as well as additional function operations such as heating sterilization operation and filter cleaning operation. The heating sterilization operation is an operation that sterilizes the indoor heat exchanger in the indoor unit 21 by heating it to 55 degrees or more. The filter cleaning operation is an operation that cleans the filter in the indoor unit 21. The recommendation unit 21D0 sends, for example, a signal recommending the execution of the operation to the user's terminal device to recommend the execution of the selected operation to the user. The recommendation unit 21D0 sends a recommendation of various air conditioning operations to the user's terminal device 7 via the communication adapter 3, the router 4, the communication network 8, and the router 4.

The behavior prediction unit 21D2 predicts the behavior of the user in the air-conditioned space based on the user presence/absence prediction result of the presence/absence prediction unit 34E described later and the indoor use information. The recommendation unit 21D0 then transmits the contents of the air-conditioning operation recommended to the user, such as the setting of the air-conditioning operation start time or the air-conditioning operation stop time, based on the behavior prediction result of the behavior prediction unit 21D2, to the user's terminal device 7. The recommended air-conditioning operation contents include, for example, the contents recommending the setting of the timer time for starting the air-conditioning operation according to the wake-up time, the return home time, the time going out, and the time going to bed. The recommendation unit 21D0 transmits a signal to the user's terminal device 7 to recommend an additional function operation different from the air-conditioning operation based on the behavior prediction result of the behavior prediction unit 21D2. The contents recommending the additional function operation include, for example, the contents proposing the timer time for starting the heat sterilization operation.

The extraction unit 21D1 extracts time periods when additional function operation, which is different from normal air conditioning operation, can be performed from the 24-hour presence/absence prediction results in the air-conditioned space. The time periods when additional function operation can be performed are, for example, absence periods of 5 hours or more. The absence periods are the absence periods of time when the user is absent in the air-conditioned space extracted from the 24-hour presence/absence prediction results predicted by the presence/absence prediction unit 34E described below. The extraction unit 21D1 calculates the available operation time from the absence periods and compares the available operation time with the operation time of the additional function operation stored in advance. Then, if the available operation time is equal to or greater than the operation time, the extraction unit 21D1 extracts the absence periods as the time periods when additional function operation can be performed.

The communication adapter 3 has a communication function for connecting the indoor unit 21 in the air conditioner 2 to the router 4 by wireless communication, and a control function for controlling the indoor unit 21 by AI (Artificial Intelligence). The communication adapter 3 is arranged for each indoor unit 21. The router 4 is an access point device that connects the communication adapter 3 to the communication network 8 by wireless communication using, for example, a WLAN (Wireless Local Area Network) or the like, and also connects the terminal device 7 to the communication network 8 by wireless communication. The communication network 8 is, for example, a communication network such as the Internet. The server device 5 has a function for generating a presence/absence pattern applied to the indoor unit 21 and a database that stores operation history data, etc. The server device 5 is arranged, for example, in a data center. The relay device 6 has a function for connecting to the communication network 8 by communication and a function for connecting to the server device 5 by communication. The relay device 6 transmits operation history data, etc. used for generating or updating the presence/absence pattern applied to the indoor unit 21 from the communication adapter 3 to the server device 5 via the communication network 8. In addition, the relay device 6 transmits the presence/absence pattern generated or updated by the server device 5 to the communication adapter 3 via the communication network 8. The relay device 6 is located, for example, in a data center or the like.

The relay device 6 has a first relay unit 6A, a second relay unit 6B, and a third relay unit 6C. The first relay unit 6A transmits various data related to the presence/absence pattern (hereinafter, referred to as operation history data) from the communication adapter 3 to the server device 5 via the communication network 8, and transmits the presence/absence pattern generated or updated by the server device 5 to the communication adapter 3 via the communication network 8. The second relay unit 6B acquires the operation conditions (such as the operation mode such as cooling/heating and the set temperature) of the indoor unit 21 set by the user using the terminal device 7 while away from home, and transmits them to the indoor unit 21. The third relay unit 6C acquires external data such as weather forecasts and calendar information (mainly holiday information) from a communication network 8 such as the Internet, and transmits the acquired external data to the server device 5. The third relay unit 6C also transmits the external data to the communication adapter 3 via the communication network 8.

<Communication adapter configuration>
FIG. 3 is a block diagram showing an example of the configuration of the communication adapter 3. The communication adapter 3 shown in FIG. 3 has a first communication unit 31, a second communication unit 32, a storage unit 33, and a CPU (Central Processing Unit) 34. The first communication unit 31 is a communication IF (Interface) such as a UART (Universal Asynchronous Receiver Transmitter) that communicatively connects the control unit 21D in the indoor unit 21 and the CPU 34. The second communication unit 32 is a communication unit such as a communication IF such as a WLAN that communicatively connects the router 4 and the CPU 34. The storage unit 33 has, for example, a ROM (Read Only Memory) or a RAM (Random Access Memory), and stores various information such as data and programs. The CPU 34 controls the entire communication adapter 3.

The memory unit 33 in the communication adapter 3 shown in FIG. 3 has a history memory 33A, a presence/absence pattern memory 33B, a prediction result memory 33C, and an external memory 33D. The history memory 33A temporarily stores operation history data acquired from the indoor unit 21. The operation history data is, for example, the detection results of the presence/absence of people in the indoor space detected by the sensor 21B every 10 minutes. The presence/absence pattern memory 33B stores the presence/absence pattern acquired from the server device 5.

The presence/absence pattern is a pattern that the server device 5 generates for each day of the week, which indicates the tendency of the user's presence/absence in the air-conditioned space, by using, for example, past detection results of the sensor 21B, for example, the presence/absence detection results for the past 30 days, day of the week information, and holiday information. In this embodiment, up to five types of presence/absence patterns are generated, and the presence/absence patterns are associated with each day of the week so that it can be determined in which presence/absence pattern the user tends to act on each day of the week. For example, on Mondays and Tuesdays, there is a tendency to act in presence/absence pattern 1, on Wednesdays and Thursdays, there is a tendency to act in presence/absence pattern 2, on Thursdays and Fridays, there is a tendency to act in presence/absence pattern 3, on Saturdays, there is a tendency to act in presence/absence pattern 4, and on Sundays, there is a tendency to act in presence/absence pattern 5. Here, the detection results of the sensor 21B for the past 30 days are used when generating the presence/absence pattern for the following reason. When generating the presence/absence pattern, it is preferable to have as many detection results of the sensor 21B as possible, because the more detection results of the sensor 21B there are, the higher the accuracy of the prediction using the presence/absence pattern will be. On the other hand, if the presence/absence pattern is generated using, for example, the detection results of the past 90 days so as to obtain many detection results of the sensor 21B, when the air conditioner 2 is installed at the beginning of summer when cooling operation is frequently performed or at the beginning of winter when heating operation is frequently performed, summer or winter will pass while the presence/absence pattern is generated, and it will be impossible to predict the user's behavior or recommend air conditioning operation based on the predicted results of the user's presence or absence, as described below, in summer or winter. Therefore, in this embodiment, the presence/absence pattern is generated using the presence/absence detection results of the sensor 21B for the past 30 days, so that the accuracy of the presence/absence pattern can be guaranteed and the user's behavior prediction and the air conditioning operation recommendation based on the predicted results of the user's presence or absence can be provided at an appropriate time. Note that the presence/absence detection results for the past 30 days are information obtained by accumulating the presence/absence detection results every 10 minutes for 30 days. In addition, in this embodiment, a case where the detection results of the sensor 21B for the past 30 days are used when generating the presence/absence pattern is illustrated, but the present invention is not limited to this. This may be changed as appropriate depending on the period between when the air conditioner 2 is installed and when it is frequently used.

The day of the week information is information on the days of the week, Mon. Tues. Wed. Thurs. Fri. Sat. Sun, and is calculated and obtained by the CPU 34. The holiday information is information that identifies the holidays among the days of the week, Mon. Tues. Wed. Thurs. Fri. Sat. Sun, and is obtained from the outside via the second communication unit 32. The reason for obtaining the holiday information from the outside is that holidays may change from year to year. The prediction result memory 33C stores the 24-hour presence/absence prediction results, which are the prediction results of the presence/absence of the user every 10 minutes for 24 hours in the air-conditioned space predicted by the presence/absence pattern, for each indoor use. The indoor use is information that identifies the use of the air-conditioned space, such as a living room or a bedroom. The indoor use is information used when predicting the user's behavior using the presence/absence prediction results. The CPU 34 can recognize the 24-hour presence/absence prediction results for each air-conditioned space by referring to the prediction result memory 33C. The external memory 33D stores external data obtained from the outside, such as the holiday information and weather forecast mentioned above.

The CPU 34 has a collection unit 34A, a transmission unit 34B, a reception unit 34C, a setting unit 34D, and a presence/absence prediction unit 34E.

The collection unit 34A acquires the detection result of the presence or absence of a person in each air-conditioned space from the indoor unit 21 at a predetermined period, for example, every 10 minutes. The air-conditioned space is, for example, an air-conditioned space such as a living room or a bedroom. The collection unit 34A collects the current detection result of the presence or absence of a person by the sensor 21B every 10 minutes in the acquired air-conditioned space. There are three types of variables for the presence or absence detection result, for example, absent, present, and indefinite. Among the presence or absence detection results, "absent" is a detection result when a person cannot be detected in the air-conditioned space. This detection result of "absent" is the second detection value. Among the presence or absence detection results, "presence" is a detection result when a person is detected in the air-conditioned space. This detection result of "presence" is the first detection value. Among the presence or absence detection results, "indefinite" is a third detection value that does not correspond to either presence or absence, that is, neither the first detection value nor the second detection value, and is not used to generate a presence or absence pattern. The collection unit 34A stores the presence/absence detection results for each air-conditioned space acquired every 10 minutes in the history memory 33A.

For example, when the transmission unit 34B stores two days' worth of presence/absence detection results in the history memory 33A, it transmits the two days' worth of presence/absence detection results stored in the history memory 33A to the server device 5 via the communication network 8. The server device 5 generates up to five types of presence/absence patterns as described above using the presence/absence detection results for the past 30 days received sequentially from the communication adapter 3. The reception unit 34C receives the presence/absence patterns for each air-conditioned space from the server device 5 via the communication network 8, and stores the received presence/absence patterns in the presence/absence pattern memory 33B. The setting unit 34D applies the stored presence/absence patterns to the presence/absence prediction unit 34E.

The presence/absence prediction unit 34E uses the current detection result of the sensor 21B, i.e., the presence/absence detection result of the sensor 21B from the time when the presence/absence is predicted until a specified time before, the current day of the week information, and the current holiday information to select a presence/absence pattern to be used for prediction from among the multiple presence/absence patterns applied by the setting unit 34D. The presence/absence prediction unit 34E predicts the presence/absence of the user in the air-conditioned space using the selected presence/absence pattern, and obtains presence/absence prediction results for 24 hours. The specified time is the time required to obtain a number of data that can ensure the accuracy of selecting the optimal presence/absence pattern from among the multiple presence/absence patterns by looking at the most recent presence/absence detection result.

Below, we will explain in detail how to select the presence/absence pattern to be used for prediction and how to predict the user's presence/absence using the selected presence/absence pattern. In the following explanation, we will explain a case where the user's presence/absence is predicted every day at 8:00, and the user's presence/absence is predicted for 24 hours from 8:00 on the current day to 8:00 on the next day. In this embodiment, the 24-hour prediction is divided into two periods: 1) 8:00 on the current day to midnight on the next day, and 2) midnight on the next day to 8:00 on the next day, and these are combined to form the 24-hour prediction result.

<1) Predicting user presence/absence from 8:00 on the current day to 0:00 on the next day>
First, when the time to predict the user's presence or absence is, for example, 8:00 on the current day, the presence/absence prediction unit 34E obtains the detection result of the user's presence or absence detected by the sensor 21B a predetermined time before the prediction time, for example, from 21:00 on the day before the prediction date to 8:00 on the current day. Next, the presence/absence prediction unit 34E compares multiple presence/absence patterns and determines whether there is a difference between each presence/absence pattern. Specifically, it determines whether the difference between the presence/absence patterns is equal to or greater than a predetermined value. More specifically, it compares the presence/absence of the user every 10 minutes from 0:00 to 8:00 in each presence/absence pattern, and if the number of points where the presence/absence differs (hereinafter referred to as "time periods") is less than a predetermined value, for example, 10, it determines that the difference between each presence/absence pattern is within an acceptable range (there is no difference between the presence/absence patterns from 0:00 to 8:00). On the other hand, if there are, for example, 10 or more time periods between 0:00 and 8:00 where the user's presence/absence varies every 10 minutes, it is determined that the difference between each presence/absence pattern exceeds the acceptable range (there is a difference in the presence/absence patterns from 0:00 to 8:00).

Next, the presence/absence prediction unit 34E selects the presence/absence pattern to be used for prediction based on the comparison result of each of the above-mentioned presence/absence patterns. If the difference between each presence/absence pattern is less than a predetermined value (if there is no difference in the presence/absence patterns from 0:00 to 8:00), it selects the presence/absence pattern associated with the day of the week to be predicted. If the difference between each presence/absence pattern is equal to or greater than a predetermined value (if there is a difference in the presence/absence patterns from 0:00 to 8:00), it compares the user's presence/absence detection result acquired from 0:00 to 8:00 with the presence/absence from 0:00 to 8:00 in each presence/absence pattern, and selects the presence/absence pattern that is closest to the detection result. Then, the presence/absence prediction unit 34E extracts the presence/absence from 8:00 to 0:00 in the above-mentioned selected presence/absence pattern as the prediction result of the user's presence/absence from 8:00 on the day to 0:00 on the next day. In this way, by associating each presence/absence pattern with day-of-the-week information and predicting the user's presence/absence based on the results of comparing each presence/absence pattern, it is possible to accurately predict the user's presence/absence while reducing the number of generated presence/absence patterns to less than the number of days of the week.

Ideally, a presence/absence pattern would be generated for each day of the week, and the presence/absence pattern to be used would be selected according to the day of the week on which the user's presence or absence is predicted. This is because the more presence/absence patterns are generated for each day of the week, the more improved the accuracy of the prediction can be expected. However, the more presence/absence patterns are increased, the greater the amount of communication between the communication adapter 3 and the server device 5, and the greater the memory capacity required by the communication adapter, placing a greater load on the air conditioning system 1.

Therefore, in this embodiment, as mentioned above, there are a maximum of five types of presence/absence patterns, and the same pattern is applied to days of the week for which the predicted presence/absence patterns can be considered to be the same. For example, presence/absence pattern 1 is applied to Monday and Tuesday, presence/absence pattern 2 is applied to Wednesday and Thursday, presence/absence pattern 3 is applied to Thursday and Friday, presence/absence pattern 4 is applied to Saturday, and presence/absence pattern 5 is applied to Sunday. However, if presence/absence patterns are generated in this way so that they can be applied to multiple days of the week, there is a risk that the accuracy of the prediction will decrease compared to when a presence/absence pattern is generated for each day of the week and these are used to predict the user's presence/absence.

To deal with such a situation, in this embodiment, the method of selecting the presence/absence pattern to be used to predict the user's presence/absence is changed depending on the comparison result of each presence/absence pattern. If the difference between each presence/absence pattern is less than a predetermined value, it is not possible to determine which presence/absence pattern to use using the user's presence/absence information acquired up to the time of prediction (8:00), so the accuracy of the prediction will not decrease if the presence/absence pattern that matches the day of the week to be predicted is selected. Also, if the difference between each presence/absence pattern is equal to or greater than a predetermined value, it is possible to distinguish between each presence/absence pattern, so the acquired user's presence/absence information is compared with the predicted results of each presence/absence pattern, and the presence/absence pattern that is closest to the detection result is selected, thereby ensuring the accuracy of the prediction.

<2) Prediction of user presence/absence from 0:00 to 8:00 the next day>
First, the presence/absence prediction unit 34E reads out the day of the week following the day for which the user's presence or absence is predicted from the external memory 33D. Next, the presence/absence prediction unit 34E selects a presence/absence pattern corresponding to the read-out day of the week from among a plurality of presence/absence patterns. Then, the presence/absence prediction unit 34E extracts a prediction result of the user's presence or absence from 0:00 to 8:00 the following day from the selected presence/absence pattern.

At 8:00 on the day when the user's presence or absence is predicted, unlike 1) when predicting the user's presence or absence from 8:00 on the day to midnight the next day, there are no detection results of the user's presence or absence detected by sensor 21B after 8:00 on the day. Therefore, when predicting the user's presence or absence from midnight to 8:00 the next day, a presence or absence pattern to be used for prediction is selected based on the day of the week of the next day, and the selected presence or absence pattern is used to predict the user's presence or absence from midnight to 8:00 the next day.

Then, the presence/absence prediction unit 34E combines the user presence/absence prediction result from 8:00 on the current day to 0:00 on the next day obtained in 1) with the user presence/absence prediction result from 0:00 on the next day to 8:00 on the next day obtained in 2) to predict the user presence/absence for 24 hours from 8:00 on the current day to 8:00 on the next day, and outputs the predicted result to the prediction result memory 33C as the presence/absence prediction result for the 24 hours. The prediction result memory 33C stores the presence/absence prediction results for the 24 hours. When the predicted time period includes a holiday, the presence/absence prediction unit 34E obtains the presence/absence prediction result for the 24 hours in the air-conditioned space by regarding the time period as the same as a holiday. Furthermore, the presence/absence prediction unit 34E excludes the presence/absence detection result of "indeterminate" (third detection value) from the presence/absence detection results, which are the detection results of the sensor 21B used when predicting the presence/absence of the user in the air-conditioned space. In other words, "undefined" presence/absence detection results are excluded and are not used to generate or update the presence/absence pattern, which improves the accuracy of predictions based on the generated or updated presence/absence pattern.

The presence/absence prediction unit 34E may predict the presence/absence of the user in the air-conditioned space for 24 hours from the predetermined time (hereinafter referred to as the "second predetermined period"), for example, at 8:00 and 20:00 every day as the predetermined time when the presence/absence is predicted. Specifically, the prediction unit 34E obtains a 24-hour presence/absence prediction result that is a prediction result of the presence/absence of the user. The reason for setting 8:00 as the predetermined time is that it is considered that the user's behavior pattern thereafter is often going out → (absent) → returning home, so that the additional function operation when absent and the start of operation when returning home are recommended. The reason for setting 20:00 as the predetermined time is that it is considered that the user's behavior pattern thereafter is often going to bed → waking up, so that the sleep operation and the start of operation when waking up are recommended. The sleep operation is an air-conditioning operation that makes the indoor temperature in the bedroom comfortable in order to provide the user with a comfortable sleeping environment. In addition, the presence/absence prediction unit 34E improves prediction accuracy by obtaining a 24-hour presence/absence prediction result from each of the above-mentioned specified times every half day. The 24-hour presence/absence prediction result is a prediction result of the presence/absence of a user in an air-conditioned space for every 10 minutes (hereinafter referred to as the "third specified period"). FIG. 4 is an explanatory diagram showing an example of a 24-hour presence/absence prediction result. The presence/absence prediction result shown in FIG. 4 is a prediction result of the presence/absence for every 10 minutes from the specified time until 24 hours later for each air-conditioned space. The data indicating the presence/absence prediction result is "1" if the user is present and "0" if the user is absent.

Next, the behavior prediction unit 21D2 predicts the user's daily behavior in the living space (e.g., wake-up time, time to go out, time to return home, time to go to bed) based on the prediction result of the user's presence or absence in the air-conditioned space and the indoor use information of the air-conditioned space (living room or bedroom). The recommendation unit 21D0 then selects an operation to be recommended to the user based on the behavior predicted by the behavior prediction unit 21D2. Furthermore, the recommendation unit 21D0 recommends to the user the execution of an operation to be recommended to the user. Below, a method will be described in which, for example, when the indoor unit 21 of the air conditioner 2 is installed in a living room or a bedroom, the control unit 21D of the indoor unit 21 predicts the user's daily behavior in the living space (e.g., wake-up time, time to go out, time to return home, time to go to bed) using the prediction result of the user's presence or absence in the living room or the bedroom, selects an operation to be recommended, and recommends the user to execute the operation to be recommended.

For example, the behavior prediction unit 21D2 of the control unit 21D in the indoor unit 21 in the living room uses the user's wake-up time (hereinafter referred to as the "first time period") and a period used as a criterion for determining whether the user is absent from the living room (hereinafter referred to as the "fourth predetermined period") in predicting the wake-up time from the presence/absence prediction result. The first time period is a time period, for example, 0:00 to 10:50, that is set in advance assuming the time when the user wakes up, leaves the bedroom, enters the living room, and starts the air conditioning operation. The fourth predetermined period is, for example, a continuous absence time of three hours or more that is used to determine that the user is asleep in the bedroom. If there is a timing in the first time period in the presence/absence prediction result in the living room that changes from absence to presence for more than the fourth predetermined period, the behavior prediction unit 21D2 predicts the timing as the wake-up time. The recommendation unit 21D0 generates a recommendation signal to start the air conditioning operation 15 minutes before the wake-up time predicted by the behavior prediction unit 21D2.

The behavior prediction unit 21D2 uses the time period when the user leaves the living room (hereinafter referred to as the "fourth time period") or the period when the user enters the bedroom (hereinafter referred to as the "fifth specified period") to predict the bedtime from the presence/absence prediction result. The behavior prediction unit 21D2 judges whether or not there is a timing in the fourth time period in the presence/absence prediction result at which the user changes from being present in the living room to being absent for a fifth specified period or more. The fourth time period is a time period, for example, 17:00 to 23:50, which is set in advance assuming the time at which the user stops the air conditioning operation in the living room and leaves the living room at night. The fifth specified period is, for example, a continuous absence period of three hours or more, which is used to determine that the user is asleep. If there is a timing in the fourth time period in the presence/absence prediction result at which the user changes from being present in the living room to being absent for a fifth specified period or more, the behavior prediction unit 21D2 predicts the timing at which the user leaves the living room as the bedtime. Then, when the recommendation unit 21D0 predicts the bedtime, it transmits a recommendation signal to the user's terminal device 7 to start air conditioning operation 15 minutes before the wake-up time between one hour before the bedtime and the bedtime. Furthermore, when there is no timing in the fourth time period in the presence/absence prediction result where the presence changes to absence for a fifth predetermined period or more, the recommendation unit 21D0 transmits a recommendation signal to start air conditioning operation 15 minutes before the wake-up time between 18:00 and 19:00 to the user's terminal device 7.

In addition, the behavior prediction unit 21D2 uses the user's time zone for returning home (hereinafter referred to as the "second time zone") and a fourth predetermined period in predicting the time of returning home from the presence/absence prediction result. The second time zone is a time zone, for example, 13:00 to 22:50, that is set in advance assuming the time when the user returns home, for example, enters the living room, and starts the air conditioning operation. If there is a timing in the second time zone in the presence/absence prediction result where the user changes from being absent for a fourth predetermined period or more to being present, the behavior prediction unit 21D2 predicts the timing as the time of returning home. The recommendation unit 21D0 generates a recommendation signal to start the air conditioning operation 15 minutes before the time of returning home predicted by the behavior prediction unit 21D2. Then, the recommendation unit 21D0 transmits a recommendation signal to start the air conditioning operation 15 minutes before the time of returning home to the user's terminal device 7 between 11:00 and 12:00.

In addition, the behavior prediction unit 21D2 uses the time period in which the user will be out (hereinafter, the "third time period") and the period used as a criterion for determining whether the user is out or not (hereinafter, the "sixth predetermined period") when predicting the time of leaving from the presence/absence prediction result. The third time period is a time period, for example, 8:00 to 16:50, that is set in advance assuming the time when the air conditioning operation will be stopped for a long time due to the user going out. The sixth predetermined period is, for example, a continuous absence time of 5 hours or more that is used to determine that the user is out. If there is a timing during the third time period in the presence/absence prediction result when the presence changes from being in the room to being absent for a sixth predetermined period or more, the behavior prediction unit 21D2 predicts the timing as the time of leaving. Then, the recommendation unit 21D0 generates a recommendation signal to perform cooling or dehumidification operation before the predicted absence time and to start heating sterilization operation one hour after the time of leaving when three days or more have passed since the previous heating sterilization operation was performed. The recommendation unit 21D0 then sends a recommendation signal to the user's terminal device 7 to start the thermal sterilization operation.

In addition, the behavior prediction unit 21D2 uses the time period in which the user will be out (hereinafter, the "third time period") and a period used as a criterion for determining whether the user is out (hereinafter, the "seventh specified period") when predicting the time of leaving from the presence/absence prediction result. The seventh specified period is, for example, a continuous absence time of one hour or more and less than five hours used to determine that the user is out. If there is a timing in the third time period in the presence/absence prediction result where the user changes from being present to being absent for the seventh specified period or more, the behavior prediction unit 21D2 predicts the timing as the time of leaving. Then, if the cumulative operation time of the indoor unit 21 is 200 hours or more, the recommendation unit 21D0 generates a recommendation signal to start a filter cleaning operation one hour after the time of leaving. Then, the recommendation unit 21D0 transmits a recommendation signal to start a filter cleaning operation to the user's terminal device 7.

The above-mentioned heating sterilization operation and filter cleaning operation may be recommended to the user after the user leaves the living room to go to sleep. In this case, if there is a continuous absence of a sixth predetermined period or more after the time when the user leaves the living room in the fourth time period, the heating sterilization operation is recommended, and if there is a continuous absence of a seventh predetermined period, the filter cleaning operation is recommended.

For example, the behavior prediction unit 21D2 of the indoor unit 21 in the bedroom uses a fourth time period, which is the time period when the user enters the bedroom, and a fifth specified period, which is the period during which the user enters the bedroom, to predict the bedtime from the presence/absence prediction result. The behavior prediction unit 21D2 determines whether or not there is a timing in the fourth time period in the presence/absence prediction result at which the user changes from being absent in the bedroom for more than the fifth specified period to being present in the bedroom. If there is a timing in the fourth time period in the presence/absence prediction result at which the user changes from being absent in the bedroom for more than the fifth specified period to being present, the behavior prediction unit 21D2 predicts that the time at which the user will be present in the bedroom is bedtime. Then, if the recommendation unit 21D0 predicts that the timing is bedtime, it generates a recommendation signal to start air conditioning operation at bedtime. The recommendation unit 21D0 transmits a recommendation signal to start air conditioning operation at bedtime to the user's terminal device 7 during the period from one hour before bedtime to bedtime.

For example, the behavior prediction unit 21D2 of the control unit 21D in the indoor unit 21 in the bedroom predicts the departure time from the presence/absence prediction result using the third time period, which is the time period when the user leaves the bedroom, and the sixth predetermined period, which is the period used as a criterion for determining whether the user has left the bedroom. The behavior prediction unit 21D2 predicts the timing as the bedroom departure time when there is a timing in the third time period in the presence/absence prediction result when the user changes from being present in the bedroom to being absent for more than the sixth predetermined period. The recommendation unit 21D0 generates a recommendation signal to start the heating sterilization operation one hour after the departure time when the cooling operation or dehumidification operation is performed before the departure time and three days or more have passed since the previous heating sterilization operation. The recommendation unit 21D0 then transmits a recommendation signal to start the heating sterilization operation to the user's terminal device 7. The departure time when the air-conditioned space is a bedroom corresponds to the departure time when the air-conditioned space is a living room.

When predicting the departure time (wake-up time) from the presence/absence prediction result, the behavior prediction unit 21D2 uses a third time period, which is the time period when the user leaves the bedroom, and a seventh specified period, which is a period used as a criterion for determining whether the user has left the bedroom. If there is a timing in the third time period in the presence/absence prediction result in the bedroom where the user changes from being present to being absent for more than the seventh specified period, the behavior prediction unit 21D2 predicts the timing as the departure time. If the cumulative operation time of the indoor unit 21 is 200 hours or more, the recommendation unit 21D0 generates a recommendation signal to start a filter cleaning operation one hour after the departure time. The recommendation unit 21D0 then transmits a recommendation signal to start a filter cleaning operation to the user's terminal device 7.

<Configuration of Terminal Device>
The terminal device 7 is a communication terminal such as a smartphone of, for example, a user who is to be an administrator among multiple users who use the air conditioning system 1. The administrator is a user who has the authority to check and execute updates and has been registered in the air conditioning system 1. Fig. 5 is a block diagram showing an example of the configuration of the terminal device 7.

The terminal device 7 shown in FIG. 5 has a communication unit 41, an operation unit 42, a display unit 43, a memory unit 44, and a CPU 45. The communication unit 41 communicates with the communication adapter 3 via the communication network 8 and the router 4. The operation unit 42 is an input interface for inputting various commands. The display unit 43 is an output interface for displaying various information. The memory unit 44 stores various information. The CPU 45 controls the entire terminal device 7. The operation unit 42 has a change unit 42A that changes the second timer time, which is the recommended time on the recommendation screen shown in FIG. 9B described later, in response to a specified operation.

<Configuration of Server Device>
Fig. 6 is a block diagram showing an example of the configuration of the server device 5. The server device 5 shown in Fig. 6 has a communication unit 51, a storage unit 52, and a CPU 53. The communication unit 51 is a communication IF that communicatively connects the relay device 6 and the CPU 53. The storage unit 52 has, for example, a hard disk drive (HDD), a ROM, a RAM, etc., and stores various information such as data and programs. The CPU 53 controls the entire server device 5.

The memory unit 52 in the server device 5 shown in FIG. 6 has a history data memory 52A and a pattern memory unit 52B. The history data memory 52A stores operation history data such as two days' worth of presence/absence detection results in the air-conditioned space received from the communication adapter 3. The pattern memory unit 52B stores the presence/absence pattern generated by the server device 5, and updates the generated presence/absence pattern using the acquired data and stores the updated presence/absence pattern.

The CPU 53 in the server device 5 has a receiving unit 53A, an acquiring unit 53B, a generating unit 53C, and a transmitting unit 53D.

The receiver 53A is connected to the communication adapters 3 of multiple indoor units 21 and receives two days' worth of presence/absence detection results for each air-conditioned space from the communication adapter 3 via the router 4, communication network 8, and relay device 6, and stores the received two days' worth of presence/absence detection results in the history data memory 52A. The receiver 53A receives day of the week information and holiday information from the communication adapter 3. The day of the week information may be calculated and obtained by the CPU 53 of the server device 5, or the holiday information may be directly obtained by the server device 5 from outside. The acquirer 53B acquires the day of the week information and holiday information received by the receiver 53A. The acquirer 53B acquires the day of the week information and holiday information received by the receiver 53A.

Figure 7 is an explanatory diagram showing an example of data used to generate a presence/absence pattern. The data used to generate a presence/absence pattern includes presence/absence detection results as sensor data, day of the week information as day of the week data, and holiday information as holiday data. As mentioned above, the presence/absence detection results are the detection results of the presence/absence of people by sensor 21B in the air-conditioned space every 10 minutes. Also, as mentioned above, the presence/absence detection results of "indefinite" are not used to generate or update a presence/absence pattern.

The generation unit 53C uses the presence/absence detection results shown in FIG. 7 for a first predetermined period stored in the history data memory 52A, for example, 30 days of past detection results, day of the week information, and holiday information, to generate a presence/absence pattern of users in the air-conditioned space of the indoor unit 21. The generation unit 53C stores the generated presence/absence pattern in the pattern storage unit 52B. If a holiday is included in the time period of the presence/absence detection results, the generation unit 53C considers the time period to be the same as a holiday. After storing the presence/absence pattern in the pattern storage unit 52B, the generation unit 53C updates the presence/absence pattern stored in the pattern storage unit 52B using the presence/absence detection results for, for example, 6 days unused for generation in the history data memory 52A, and stores the updated presence/absence pattern in the pattern storage unit 52B.

When the air conditioner 2 is installed in the living room, for example, the generation unit 53C extracts a presence/absence detection result for a weekday, for example, Monday (excluding holidays), from the presence/absence detection result for the living room stored in the history data memory 52A. Furthermore, the generation unit 53C extracts a presence/absence detection result other than "undefined" from the above-extracted presence/absence detection result for Monday, and generates a presence/absence pattern that predicts the presence/absence of a person in the living room on Monday based on the extracted presence/absence detection result for the living room. Furthermore, when the air conditioner 2 is installed in the bedroom, for example, the generation unit 53C extracts a presence/absence detection result for a weekday, for example, Monday (excluding holidays), from the presence/absence detection result for the bedroom stored in the history data memory 52A. Furthermore, the generation unit 53C extracts a presence/absence detection result other than "undefined" from the above-extracted presence/absence detection result for Monday, and generates a presence/absence pattern that predicts the presence/absence of a person in the bedroom on Monday based on the extracted presence/absence detection result for the bedroom.

The generation unit 53C also extracts the presence/absence detection results for holidays and Sundays from the presence/absence detection results for the living room stored in the history data memory 52A. Furthermore, the generation unit 53C extracts the presence/absence detection results other than "undefined" from the extracted presence/absence detection results for holidays and Sundays, and generates a presence/absence pattern that predicts the presence/absence of a person in the living room on Sunday based on the extracted presence/absence detection results for the living room. The generation unit 53C also extracts the presence/absence detection results for holidays and Sundays from the presence/absence detection results for the bedroom stored in the history data memory 52A. Furthermore, the generation unit 53C extracts the presence/absence detection results other than "undefined" from the extracted presence/absence detection results for holidays and Sundays, and generates a presence/absence pattern that predicts the presence/absence of a person in the bedroom on Sunday based on the extracted presence/absence detection results for the bedroom.

In other words, the generation unit 53C generates a presence/absence pattern for each day of the week in the air-conditioned space in which the indoor unit 21 is installed. For ease of explanation, an example has been given of generating a presence/absence pattern for each day of the week, but for example, Monday through Friday excluding holidays may be set as weekdays, and a presence/absence pattern for each air-conditioned space on weekdays may be generated, and holidays, Saturdays, and Sundays may be set as holidays, and a presence/absence pattern for each air-conditioned space on holidays may be generated. In addition, while holidays, Saturdays, and Sundays have been given as examples of holidays, this is not limited to these, and for example, Tuesdays may be set as holidays regardless of holidays on the calendar, and changes can be made as appropriate.

Figure 8 is an explanatory diagram showing an example of a generated user presence/absence pattern. Pattern 1 of the presence/absence patterns shown in Figure 8 is a presence/absence pattern that indicates the presence/absence of a user in the air-conditioned space on Mondays and Tuesdays. Although not shown, the presence/absence pattern of a user in the air-conditioned space from Wednesdays to Saturdays other than holidays is also predicted. Pattern 2 is a presence/absence pattern that indicates the presence/absence of a user in the air-conditioned space on Sundays and holidays.

The generation unit 53C generates or updates a presence/absence pattern for each day of the week for each air-conditioned space based on the presence/absence detection results, the day of the week information, and the holiday information, and stores the generated or updated presence/absence pattern in the pattern storage unit 52B. The transmission unit 53D transmits the presence/absence pattern for each day of the week for each air-conditioned space stored in the pattern storage unit 52B to the communication adapter 3 via the relay device 6, the communication network 8, and the router 4.

<Display form of terminal device>
FIG. 9A is an explanatory diagram showing an example of a timer setting screen 70 of the terminal device 7. The timer setting screen 70 shown in FIG. 9A displays an operation mode display section 71 that displays the operation mode currently being operated, a set temperature display section 72 that displays the current set temperature, a set air volume display section 73 that displays the current set air volume, a first timer display section 74 that displays the contents of the reserved operation set by the user, and a second timer display section 75 that displays the contents of the reserved operation (AI operation proposal) recommended by the air conditioner 2. The operation mode display section 71 displays the currently operating mode, such as normal air conditioning operation such as cooling, heating, blowing, and drying, and additional function operation such as filter cleaning operation and heat sterilization operation. The first timer display section 74 is a first display section that displays timer times such as the start time and end time of the air conditioning operation according to the setting operation of the user. The second timer display section 75 is a second display section that displays timer times such as the start time of the air conditioning operation recommended by the recommendation section 21D0 of the air conditioner 2. By displaying the first timer display section 74 and the second timer display section 75 separately on the timer setting screen 70, the user can easily distinguish between the operation of the timer set by the user through the first timer display section 74 and the AI operation suggestion through the second timer display section 75. As a result, the user can prevent erroneous operation such as accidentally erasing the AI operation suggestion, compared to when the two are displayed without distinction.

9B is an explanatory diagram showing an example of a recommendation screen 80 of the terminal device 7. When the second timer display section 75 on the timer setting screen 70 shown in FIG. 9A is selected by the user, the terminal device 7 displays the recommendation screen 80 shown in FIG. 9B. The recommendation screen 80 displays a reservation execution selection section 81, a second timer time display section 82 that displays the timer setting time of the AI operation proposal, a recommended content display section 83 that displays a sentence explaining the content of the AI operation proposal, and a recommended setting selection section 84. The reservation execution selection section 81 is an operation section that instructs the air conditioner 2 whether to execute a reservation. The second timer time display section 82 displays the timer time recommended in the recommendation section 21D0 of the air conditioner 2. The recommended content display section 83 pops up the recommended content recommended in the recommendation section 21D0. The recommended content is, for example, "It looks like it's going to be cold tomorrow morning. Why don't you set an operation reservation so that it will be comfortable at 6:00 a.m." The recommended settings selection unit 84 is an operation unit that instructs the air conditioner 2 whether or not to set the recommended content.

Figure 9C is an explanatory diagram showing an example of a recommended time change screen 90 of the terminal device 7. When the second timer time display section 82 on the recommendation screen 80 shown in Figure 9B is selected by the user, the terminal device 7 displays the change screen 90 shown in Figure 9C. The second timer time display section 82 corresponds to the change section 42A, and displays the change screen 90 in response to the user's selection operation. The terminal device 7 can freely adjust the second timer time in response to the user's change operation of the second timer time 91 on the change screen 90. As a result, the user can freely adjust the recommended timer time for AI operation.

For ease of explanation, the timer setting screen 70 shown in FIG. 9A, the recommendation screen 80 shown in FIG. 9B, and the change screen 90 shown in FIG. 9C are shown as examples of screens displayed on the terminal device 7, but the timer setting screen 70, the recommendation screen 80, and the change screen 90 may each be displayed on the remote control 23. In addition, in the above explanation, the first timer display unit 74 and the second timer display unit 75 are provided in advance on the terminal device 7, but for example, when an air conditioner 2 that originally cannot make AI operation suggestions becomes able to make AI operation suggestions due to a program update or the like, one of the multiple first timer display units 74 may be changed to the second timer display unit 75.

<Recommendations for operating additional functions>
Next, the recommendation of the additional function operation from the air conditioner 2 to the user, which is different from the normal air conditioning operation, will be described. The additional function operation in this embodiment includes a heat sterilization operation and a filter cleaning operation. FIG. 10 is an explanatory diagram showing an example of a timing chart of each process during the heat sterilization operation. The recommendation unit 21D0 in the air conditioner 2 obtains the 24-hour presence/absence prediction result, which is the prediction result of the presence/absence prediction unit 34E, and detects a sixth predetermined period, for example, a time when absence of 5 hours or more is predicted from this prediction result. Then, the recommendation unit 21D0 recommends the heat sterilization operation to the user only when the cooling operation or dehumidification operation is performed immediately before (for example, 1 hour before) the detected absence time of 5 hours or more, and 3 days or more have passed since the previous heat sterilization operation was performed. The air conditioner 2 notifies the terminal device 7 of the recommendation of the heat sterilization operation at the time when the cooling operation or dehumidification operation is stopped. If a user who has received the notification recommending the heat sterilization operation does not refuse the execution of the heat sterilization operation, the air conditioner 2 will execute the heat sterilization operation during the detected absence period of 5 hours or more. The heat sterilization operation is executed through the following procedure. First, the air conditioner 2 executes a monitoring operation, which will be described later, from the point at which the cooling operation or dehumidification operation is stopped until the first hour of the absence period of 5 hours or more. After the monitoring operation is completed, the heat sterilization operation is executed for 1 hour, and the remaining 3 hours are used as a cooling period to reduce the indoor temperature and indoor humidity that have increased due to the heat sterilization operation.

The above-mentioned monitoring operation is an operation that monitors whether or not a person is actually present in the air-conditioned space using the presence/absence detection result of sensor 21B. When the additional function operation is heat sterilization operation, the monitoring time is the time from when the cooling operation or dehumidification operation is stopped to the time when the heat sterilization operation is performed, and a monitoring operation is performed to monitor the presence or absence of a user in the air-conditioned space during this monitoring time. If a person is detected in the air-conditioned space during the monitoring operation, the air conditioner 2 does not perform the heat sterilization operation and notifies the user's terminal device 7 that this heat sterilization operation will not be performed. Note that FIG. 10 shows a case where the cooling operation or dehumidification operation is stopped at the time when people are absent and the monitoring time begins.

As mentioned above, the heating sterilization operation is an operation that heats the indoor heat exchanger of the indoor unit 21 to 55 degrees to sterilize the air. The heating sterilization operation causes the indoor temperature and humidity in the air-conditioned space to rise. Therefore, after the heating sterilization operation is performed, a cooling period is set up to allow the indoor temperature and humidity increased by the heating sterilization operation to fall. The cooling period is a preparation time that should be set up after the heating sterilization operation ends and before normal air-conditioning operation is performed. As a result, it is possible to avoid situations where the heating sterilization operation causes discomfort to users in the air-conditioned space.

Figure 11 is an explanatory diagram showing an example of a timing chart of each process during filter cleaning operation. The recommendation unit 21D0 in the air conditioner 2 recommends filter cleaning operation to the user when an absence time of 1 hour or more and less than 5 hours is detected in the 24-hour presence/absence prediction result, which is the prediction result of the presence/absence prediction unit 34E, and the accumulated operation time of the indoor unit 21 is 200 hours or more. If the user accepts the recommended filter cleaning operation, the air conditioner 2 performs a monitoring operation in the first hour of the absence time of 1 hour or more and less than 5 hours, and a filter cleaning operation in the next 20 minutes. The monitoring operation is an operation that monitors whether or not a person is actually absent in the air-conditioned space using the presence/absence detection result of the sensor 21B. Specifically, the first hour of the absence time of 1 hour or more and less than 5 hours is set as the monitoring time before starting the filter cleaning operation, and a monitoring operation is performed to monitor the presence or absence of the user in the air-conditioned space during the monitoring time. Then, during the monitoring operation, a notification is given to the terminal device 7 that a person has been detected in the air-conditioned space. As a result, it is possible to prevent situations in which filter cleaning operation causes discomfort to users in the air-conditioned space.

As described above, in the air conditioning system 1 of this embodiment, the execution of additional function operations that may cause discomfort to the user if executed while the user is present in the room is recommended according to the predicted length of time the user will be absent. As a result, the indoor heat exchanger of the indoor unit 21 can be sterilized and the filter cleaned automatically without causing discomfort to the user.

<Generation of occupancy patterns in air conditioning systems>
Next, the generation of the presence/absence pattern in the air conditioning system 1 of this embodiment will be described. FIG. 12A is a flowchart showing an example of the processing operation of the CPU 53 of the server device 5 related to the generation process for generating the presence/absence pattern. The generation process is a process for generating the presence/absence pattern for the first time after the air conditioner 2 is installed in the air-conditioned space. In FIG. 12A, the receiving unit 53A in the CPU 53 of the server device 5 periodically communicates with the communication adapter 3 at 0:00 every day, for example, and judges whether or not the presence/absence detection results for two days for each air-conditioned space have been received from the communication adapter 3 (step S11). Note that the communication adapter 3 stores the presence/absence detection results for two days in the history memory 33A until they are obtained. When the receiving unit 53A receives the presence/absence detection results for two days (step 11: Yes), it stores the received presence/absence detection results for the two days in the history data memory 52A of the storage unit 52 (step S12). The generating unit 53C in the CPU 53 judges whether or not the presence/absence detection results for 30 days have been stored in the history data memory 52A (step S13). When the generation unit 53C has stored the presence/absence detection results for 30 days (step S13: Yes), it generates a presence/absence pattern for each day of the week for each air-conditioned space based on the stored presence/absence detection results, day of the week information, and holiday information (step S14). When acquiring the presence/absence detection results for two days, the acquisition unit 53B in the CPU 53 also acquires the day of the week information and holiday information of the detection date of the presence/absence detection results for the two days, or links the day of the week information and holiday information acquired by the server device 5 to the presence/absence detection results for the two days acquired. In addition, the generation unit 53C illustrates an example in which a presence/absence pattern is generated for each day of the week, but two presence/absence patterns for holidays and weekdays may be generated, and can be changed as appropriate.

The generation unit 53C stores the generated presence/absence pattern in the pattern storage unit 52B (step S15). The transmission unit 53D in the CPU 53 transmits the presence/absence pattern stored in the pattern storage unit 52B to the communication adapter 3 (step S16), and the processing operation of FIG. 12A ends.

If the receiving unit 53A has not received two days' worth of presence/absence detection results for each air-conditioned space in the process of step S11 (step S11: No), the process returns to step S11. Also, if the receiving unit 53A has not stored 30 days' worth of presence/absence detection results in the process of step S13 (step S13: No), the process returns to step S11.

When the CPU 53 of the server device 5 stores the presence/absence detection results for 30 days for each air-conditioned space from the communication adapter 3, it generates a presence/absence pattern for each day of the week that predicts the presence/absence of the user in the air-conditioned space based on the presence/absence detection results for 30 days for each air-conditioned space, the day of the week information, and the holiday information. The CPU 53 then transmits the generated presence/absence pattern to the communication adapter 3. As a result, the server device 5 can provide the communication adapter 3 with the presence/absence pattern for each day of the week to be used in the air-conditioned space.

Figure 12B is a flowchart showing an example of the processing operation of the CPU 53 of the server device 5 related to the update process for updating an already generated presence/absence pattern. The update process is a process for updating the contents of the presence/absence pattern stored in the pattern storage unit 52B. In Figure 12B, the receiving unit 53A communicates with the communication adapter 3 periodically, for example, at 0:00 every day, and determines whether or not it has received two days' worth of presence/absence detection results for each air-conditioned space from the communication adapter 3 (step S21). Note that the communication adapter 3 stores the presence/absence detection results for two days in the history memory 33A until they are obtained. When the receiving unit 53A receives two days' worth of presence/absence detection results for each air-conditioned space (step S21: Yes), it stores the received two days' worth of presence/absence detection results in the history data memory 52A of the storage unit 52 (step S22). The generating unit 53C determines whether or not six days' worth of presence/absence detection results unused for generation have already been stored in the history data memory 52A (step S23).

If the generation unit 53C has stored the presence/absence detection results for six days that have not been used for generation (step S23: Yes), it updates the presence/absence pattern for each day of the week for each air-conditioned space based on the stored presence/absence detection results, day of the week information, and holiday information (step S24). The generation unit 53C stores the updated presence/absence pattern for each day of the week for each air-conditioned space in the pattern storage unit 52B (step S25). The transmission unit 53D transmits the presence/absence pattern for each day of the week for each air-conditioned space stored in the pattern storage unit 52B to the communication adapter 3 (step S26). The reception unit 53A then returns to the processing of step S21 to determine whether or not it has received the presence/absence detection results for two days for each air-conditioned space.

If the receiving unit 53A has not received two days of presence/absence detection results in the process of step S21 (step S21: No), the process returns to step S21. Also, if the receiving unit 53A has not stored six days of presence/absence detection results that have not been used for generation in the process of step S23 (step S23: No), the process returns to step S21.

After generating the presence/absence pattern, the CPU 53 updates the presence/absence pattern for each day of the week for each air-conditioned space based on the presence/absence detection results for the six days, the day of the week information, and the holiday information for each air-conditioned space each time it obtains the presence/absence detection results for the six days from the communication adapter 3. The CPU 53 then transmits the updated presence/absence pattern to the communication adapter 3. As a result, the server device 5 can provide the communication adapter 3 with the latest presence/absence pattern for each day of the week to be used for the air-conditioned space.

Figure 13A is a flowchart showing an example of the processing operation of the CPU 34 of the communication adapter 3 related to the recommendation process. In Figure 13A, the presence/absence prediction unit 34E in the CPU 34 in the communication adapter 3 determines whether the current time is the predicted time (step S30). The predicted time is a predetermined time set in advance, such as 8:00 or 20:00 every day as described above. If the current time is the predicted time (step S30: Yes), the presence/absence prediction unit 34E determines whether the presence/absence detection result for a predetermined time from the current predicted time, for example, up to 24 hours ago, has been obtained from the indoor unit 21 of the air conditioner 2 (step S31). For convenience of explanation, the predetermined time is set to 24 hours, but is not limited to this, and may be, for example, a time period from 21:00 the previous day to 0:00 the current day, or from 0:00 the current day to the predicted time, and can be changed as appropriate.

When the presence/absence prediction unit 34E obtains the presence/absence detection result for the 24 hours prior to the current prediction time (step S31: Yes), it selects a presence/absence pattern to be used for the presence/absence prediction in the air-conditioned space from multiple presence/absence patterns in the air-conditioned space using the obtained presence/absence detection result and day of the week information (step S32). The presence/absence prediction unit 34E predicts the presence/absence of the user 24 hours from now using the selected presence/absence pattern (step S33). The behavior prediction unit 21D2 of the control unit 21D in the indoor unit 21 determines whether the indoor unit 21 is installed in a bedroom based on the indoor use information (step S34).

If the indoor unit 21 is not installed in a bedroom (step S34: No), the behavior prediction unit 21D2 executes the process of recommending air conditioning operation in the living room (step S35) and returns to the process of step S30. If the indoor unit 21 is not installed in a bedroom (step S34: Yes), the behavior prediction unit 21D2 executes the process of recommending air conditioning operation in the bedroom (step S36) and returns to the process of step S30. If the current time is not the predicted time (step S30: No), the presence/absence prediction unit 34E returns to the process of step S30. If the presence/absence prediction unit 34E has not obtained a presence/absence detection result for the 24 hours prior to the current predicted time (step S31: No), the process returns to the process of step S31.

The communication adapter 3 predicts the presence or absence of the user for 24 hours from the predicted time using a presence/absence pattern selected using the current detection results of the sensor and day of the week information, and then executes processing to recommend air conditioning operation in the bedroom if the indoor unit 21 is installed in a bedroom, and executes processing to recommend air conditioning operation in the living room if the indoor unit 21 is not installed in a bedroom. As a result, it is possible to recommend to the user air conditioning operation according to the user behavior prediction results described later using Figures 13B to 13D. Note that, for ease of explanation, a bedroom and a living room are given as examples of air-conditioned spaces, but they are not limited to these and may be a bathroom changing room, a toilet, etc., and can be changed as appropriate.

Next, the air conditioning operation recommendation process in the living room in step S35 of FIG. 13A will be described. FIGS. 13B and 13C are flowcharts showing an example of the processing operation of the control unit 21D of the indoor unit 21 related to the air conditioning operation recommendation process in the living room. The indoor unit 21 described in FIGS. 13B and 13C is an indoor unit installed in the living room. In FIG. 13B, the behavior prediction unit 21D2 in the control unit 21D in the indoor unit 21 determines whether or not there is a timing during the first time period (0:00 to 10:50) in the presence/absence prediction result that changes from absence in the living room to presence for a fourth predetermined period, for example, 3 hours or more (step S41). As described above, the first time period is the time period during which the user is expected to wake up, for example, leave the bedroom and enter the living room, and start the air conditioning operation.

If there is a timing in the first time period when the absence changes from a fourth predetermined period or more to a presence in the room (step S41: Yes), the behavior prediction unit 21D2 predicts the timing as the wake-up time (step S42). After predicting the wake-up time, the recommendation unit 21D0 generates a recommendation signal to start the air conditioning operation 15 minutes before the wake-up time (step S43). Next, the behavior prediction unit 21D2 determines whether there is a timing in the fourth time period (17:00 to 23:50) in the presence/absence prediction result when the presence/absence changes from a fifth predetermined period, for example, a presence in the living room to an absence of three hours or more (step S44). The fourth time period is a time period when the user is expected to stop the air conditioning operation in the living room at night and leave the living room, for example, 17:00 to 23:50, or a time period when the user is expected to start the air conditioning operation in the bedroom at night and enter the bedroom, for example, 17:00 to 23:50.

If there is a timing in the fourth time period when the absence changes to a fifth predetermined period or more (step S44: Yes), the behavior prediction unit 21D2 predicts that timing as bedtime. If the behavior prediction unit 21D2 can predict bedtime, the recommendation unit 21D0 transmits the recommendation signal generated in step S43 to the user's terminal device 7 via the communication adapter 3, router 4, communication network 8, and router 4 during the period from one hour before bedtime to the bedtime (step S45), and ends the air conditioning operation recommendation process for the living room shown in FIG. 13B. As a result, the user can easily set the second timer time for starting air conditioning operation in the living room 15 minutes before the wake-up time based on the recommendation content on the display screen of the terminal device 7.

If there is no timing in the fourth time period when the absence will change to a fifth predetermined period or longer (step S44: No), the recommendation unit 21D0 transmits the recommendation signal generated in step S43 to the user's terminal device 7 via the communication adapter 3, router 4, communication network 8, and router 4 between 18:00 and 19:00 (step S46), and ends the air conditioning operation recommendation process for the living room shown in FIG. 13B. As a result, the user can easily set the second timer time for starting air conditioning operation in the living room 15 minutes before the wake-up time based on the recommendation content on the display screen of the terminal device 7.

If there is no timing in the first time period when the user changes from being absent for a fourth predetermined period or more to being present (step S41: No), the behavior prediction unit 21D2 determines whether there is a timing in the second time period (13:00 to 22:50) in the presence/absence prediction result when the user changes from being absent in the living room for a fourth predetermined period, for example, three hours or more, to being present (step S47). Note that the second time period is the time period when the user is expected to return home, for example, enter the living room, and start the air conditioning, as described above.

If there is a timing in the second time period when the user changes from being absent in the living room to being present for a period of time equal to or greater than a fourth predetermined period (step S47: Yes), the behavior prediction unit 21D2 predicts that timing as the time of arrival home (step S48). After predicting the time of arrival home, the recommendation unit 21D0 generates a recommendation signal to start air conditioning operation 15 minutes before the time of arrival home (step S49). The recommendation unit 21D0 transmits the generated recommendation signal to the user's terminal device 7 via the communication adapter 3, router 4, communication network 8, and router 4 between 11:00 and 12:00 (step S50), and ends the process of recommending air conditioning operation in the living room shown in FIG. 13B. As a result, the user can easily set the second timer time for starting air conditioning operation in the living room 15 minutes before the time of arrival home based on the recommendation content on the display screen of the terminal device 7.

In addition, the recommendation screen 80 of the terminal device 7 that receives the recommendation signals transmitted in the processes of steps S45, S46, and S50 displays content such as the recommended content display section 83 shown in FIG. 9B, and the user can select whether to accept (set) the recommended air conditioning operation recommendation or not (not set).

If there is no timing in the second time period when the user changes from being absent for a fourth predetermined period or more to being present (step S47: No), the behavior prediction unit 21D2 proceeds to M1 shown in FIG. 13C. In M1 shown in FIG. 13C, the behavior prediction unit 21D2 determines whether there is a timing in the third time period (8:00 to 16:50) in the presence/absence prediction result when the user changes from being present in the living room to being absent for a sixth predetermined period, for example, five hours or more (step S51). The third time period is a time period when the air conditioning operation is expected to be stopped for a long period of time when the user goes out. If there is a timing in the third time period when the user changes from being present in the living room to being absent for a sixth predetermined period or more (step S51: Yes), the behavior prediction unit 21D2 predicts that timing as the time of leaving the room (step S52). After predicting that the timing is the time when the person leaves the house, the recommendation unit 21D0 determines whether cooling or dehumidification operation was performed before the time when the person was absent, and whether three or more days have passed since the previous heating and sterilization operation was performed (step S53).

If cooling or dehumidification operation was performed before the absence time and more than three days have passed since the previous heating sterilization operation (step S53: Yes), the recommendation unit 21D0 generates a recommendation signal to start heating sterilization operation one hour after the time of leaving (step S54). The recommendation unit 21D0 then transmits the generated recommendation signal to the user's terminal device 7 via the communication adapter 3, the router 4, the communication network 8, and the router 4 (step S55), and ends the air conditioning operation recommendation process in the living room shown in FIG. 13B. As a result, the user can easily set the second timer time for starting heating sterilization operation in the living room one hour after the time of leaving, based on the recommendation content on the display screen of the terminal device 7.

If the recommendation unit 21D0 determines in step S53 that cooling or dehumidification operation was performed before the absence time and more than three days have not passed since the previous heating and sterilization operation was performed (step S53: No), the recommendation unit 21D0 advances the process to M2 shown in FIG. 13B and ends the process of recommending air conditioning operation in the living room.

If there is no timing during the third time period when the presence in the living room changes to an absence for a sixth predetermined period or more (step S51: No), the behavior prediction unit 21D2 determines whether there is a timing during the third time period in the presence/absence prediction result when the presence in the living room changes to an absence for a seventh predetermined period, for example, an absence for one hour or more but less than five hours (step S56). If there is a timing during the third time period when the presence in the living room changes to an absence for a seventh predetermined period (step S56: Yes), the behavior prediction unit 21D2 predicts that timing as the time to leave the house (step S57).

After predicting that timing as the time when the user leaves the house, the recommendation unit 21D0 determines whether the cumulative operating time of the indoor unit 21 is 200 hours or more (step S58). If the cumulative operating time of the indoor unit 21 is 200 hours or more (step S58: Yes), the recommendation unit 21D0 generates a recommendation signal to start filter cleaning operation one hour after the time when the user leaves the house (step S59). The recommendation unit 21D0 returns to the process of step S55 to send the generated recommendation signal to the user's terminal device 7 via the communication adapter 3, router 4, communication network 8, and router 4. As a result, the user can easily set the second timer time for starting filter cleaning operation in the living room one hour after the time when the user leaves the house based on the recommendation content on the display screen of the terminal device 7.

Incidentally, the recommendation screen 80 of the terminal device 7 that received the recommendation signal transmitted in the processes of steps S54 and S59 displays the recommendation content in the recommendation content display section 83 (not shown). However, unlike the recommendation of air conditioning operation, when additional function operation is recommended, a message is displayed to the user informing them that the additional function operation will be performed, and the additional function operation will be performed unless the user refuses the execution of the additional function operation. If the user does not want to run the additional function operation, the recommendation content allows the user to select refusal to run the additional function operation.

If there is no timing during the third time period when the presence in the living room changes to absence within the seventh predetermined period (step S56: No), the behavior prediction unit 21D2 ends the air conditioning operation recommendation process for the living room shown in FIG. 13B. Also, if the accumulated operation time of the indoor unit 21 is not 200 hours or more (step S58: No), the recommendation unit 21D0 advances the process to M2 shown in FIG. 13B and ends the air conditioning operation recommendation process for the living room.

Next, the air conditioning operation recommendation process in the bedroom in step S36 in FIG. 13A will be described. FIG. 13D is a flowchart showing an example of the processing operation of the control unit 21D of the indoor unit 21 related to the air conditioning operation recommendation process in the bedroom. The indoor unit 21 described in FIG. 13D is an indoor unit installed in the bedroom. In FIG. 13D, the behavior prediction unit 21D2 in the control unit 21D in the indoor unit 21 determines whether there is a timing during the fourth time period (17:00 to 23:50) in the presence/absence prediction result that changes from absence in the bedroom to presence for a fifth predetermined period, for example, three hours or more (step S61). As described above, the fourth time period is a time period that is assumed to be the time when the user starts the air conditioning operation in the bedroom at night and enters the bedroom, for example, 17:00 to 23:50.

If there is a timing in the fourth time period when the user changes from being absent for a fifth predetermined period or more to being present (step S61: Yes), the behavior prediction unit 21D2 predicts that the user will enter the bedroom as bedtime (step S62). After predicting bedtime, the recommendation unit 21D0 generates a recommendation signal to start air conditioning operation at bedtime (step S63). The recommendation unit 21D0 transmits the generated recommendation signal to the user's terminal device 7 via the communication adapter 3, router 4, communication network 8, and router 4 during the period from one hour before bedtime to bedtime (step S64), and ends the processing operation related to the air conditioning operation recommendation process in the bedroom shown in FIG. 13D. As a result, the user can easily set the second timer time for starting air conditioning operation in the bedroom at bedtime based on the recommendation content on the display screen of the terminal device 7.

In addition, the recommendation screen 80 of the terminal device 7 that receives the recommendation signal transmitted in each process of step S64 displays content such as the recommended content display section 83 shown in FIG. 9B, and the user can receive (set) the recommended air conditioning operation and execute the recommended air conditioning operation.

If there is no timing in the fourth time period when the absence changes from a fifth predetermined period or more to a presence (step S61: No), the behavior prediction unit 21D2 determines whether there is a timing in the third time period (8:00 to 16:50) in the presence/absence prediction result when the presence in the bedroom changes to a sixth predetermined period, for example, an absence of five hours or more (step S65). If there is a timing in the third time period when the presence in the bedroom changes to an absence of the sixth predetermined period or more (step S65: Yes), the behavior prediction unit 21D2 predicts the timing as the time of leaving the bedroom (step S66). After predicting the timing as the time of leaving the bedroom, the recommendation unit 21D0 determines whether cooling or dehumidification operation was performed before the time of leaving the bedroom and whether three days or more have passed since the previous heating and sterilization operation was performed (step S67).

If cooling or dehumidification operation was performed before the departure time and more than three days have passed since the previous heating sterilization operation was performed (step S67: Yes), the recommendation unit 21D0 generates a recommendation signal to start heating sterilization operation one hour after the departure time (step S68). The recommendation unit 21D0 then transmits the generated recommendation signal to the user's terminal device 7 via the communication adapter 3, router 4, communication network 8, and router 4 (step S69), and ends the air conditioning operation recommendation process in the bedroom shown in FIG. 13D. As a result, the user can easily set the second timer time for starting heating sterilization operation one hour after the departure time in the bedroom based on the recommendation content on the display screen of the terminal device 7.

In addition, the recommendation screen 80 of the terminal device 7 that received the recommendation signal transmitted in each process of step S69 and step S73 displays the recommendation content in the recommendation content display section 83 (not shown). However, unlike the recommendation of air conditioning operation, when an additional function operation is recommended, a message is displayed to the user informing them that the additional function operation will be performed, and the additional function operation will be performed unless the user refuses to perform the additional function operation.

If cooling or dehumidification operation was performed before the departure time in step S67 and more than three days have passed since the previous heating and sterilization operation was performed (step S67: No), the recommendation unit 21D0 ends the air conditioning operation recommendation process for the bedroom shown in FIG. 13D.

If there is no timing during the third time period when the user changes from being present in the bedroom to being absent for a sixth specified period or longer (step S65: No), the behavior prediction unit 21D2 determines whether there is a timing during the third time period in the presence/absence prediction result when the user changes from being present in the bedroom to being absent for a seventh specified period, for example, one hour or more but less than five hours (step S70). If there is a timing during the third time period when the user changes from being present in the bedroom to being absent for a seventh specified period (step S70: Yes), the behavior prediction unit 21D2 predicts that timing as the time to leave the bedroom (step S71).

After predicting that timing as the departure time, the recommendation unit 21D0 determines whether the accumulated operating time of the indoor unit 21 is 200 hours or more (step S72). If the accumulated operating time of the indoor unit 21 is 200 hours or more (step S72: Yes), the recommendation unit 21D0 generates a recommendation signal to start filter cleaning operation one hour after the departure time (step S73). The recommendation unit 21D0 returns to the process of step S69 to send the generated recommendation signal to the user's terminal device 7 via the communication adapter 3, router 4, communication network 8, and router 4. As a result, the user can easily set the second timer time for starting filter cleaning operation one hour after the departure time in the bedroom based on the recommendation content on the display screen of the terminal device 7.

If there is no timing during the third time period when the presence in the bedroom changes to absence within the seventh predetermined period (step S70: No), the behavior prediction unit 21D2 ends the air conditioning operation recommendation process in the bedroom shown in FIG. 13D. Also, if the accumulated operation time of the indoor unit 21 is not 200 hours or more (step S72: No), the recommendation unit 21D0 ends the air conditioning operation recommendation process in the bedroom shown in FIG. 13D.

Effects of Example 1
In the communication adapter 3 of the first embodiment, one presence/absence pattern is selected from a plurality of presence/absence patterns using at least the current detection result of the sensor 21B. Furthermore, the communication adapter 3 predicts the presence/absence in the air-conditioned space for 24 hours using the selected presence/absence pattern. Furthermore, the communication adapter 3 transmits the presence/absence prediction result for the 24 hours, which is the prediction result of the presence/absence, to the air conditioner 2. The air conditioner 2 transmits a recommendation signal recommending various air conditioning operations to the user's terminal device 7 via the communication adapter 3 based on the acquired 24-hour presence/absence prediction result. As a result, the user can achieve air conditioning operation in accordance with their life rhythm by implementing the recommendation content of the recommendation signal, thereby obtaining a comfortable air-conditioned space.

The server device 5 generates a presence/absence pattern using the past presence/absence detection results, day of the week information, and holiday information detected by the sensor 21B over a first predetermined period, for example, 30 days. As a result, a highly accurate presence/absence pattern in the air-conditioned space can be obtained.

The communication adapter 3 uses the selected presence/absence pattern to predict, at a specified time, for example 8:00 or 20:00 every day, a 24-hour presence/absence prediction that indicates the presence or absence of a user in the air-conditioned space for a second specified period, for example up to 24 hours. Furthermore, the communication adapter 3 transmits the 24-hour presence/absence prediction result that is the presence/absence prediction result to the air conditioner 2. The air conditioner 2 then transmits a recommendation signal recommending various air conditioning operations to the user's terminal device 7 via the communication adapter 3 based on the acquired 24-hour presence/absence prediction result. As a result, the user can receive recommendations for air conditioning operations that are in line with their lifestyle rhythm, using the presence/absence prediction result for the period up to 24 hours from the specified time.

Using the selected presence/absence pattern, the communication adapter 3 predicts the presence/absence of the user in the air-conditioned space for a third predetermined period, for example, every 10 minutes for 24 hours. Furthermore, the communication adapter 3 transmits the presence/absence prediction result for the 24 hours, which is the prediction result of presence/absence, to the air conditioner 2. The air conditioner 2 then transmits a recommendation signal recommending various air conditioning operations to the user's terminal device 7 via the communication adapter 3 based on the obtained 24-hour presence/absence prediction result. As a result, the user can receive recommendations for air conditioning operations that are in line with their lifestyle rhythm, using the presence/absence prediction result for every 10 minutes for 24 hours.

The air conditioner 2 transmits a recommendation signal to the user's terminal device 7, recommending the setting of the start time or stop time of air conditioning operation, based on the 24-hour presence/absence prediction results in the air-conditioned space predicted by the communication adapter 3. As a result, the user can realize a comfortable air-conditioned space that matches the rhythm of their life by following the recommendation content according to the recommendation signal, without having to search for an operating time period for air conditioning operation such as sleep mode.

The air conditioner 2 transmits a recommendation signal to the user's terminal device 7 recommending additional function operation other than air conditioning operation based on the predicted 24-hour presence/absence prediction results in the air-conditioned space. As a result, the user can realize additional function operation in accordance with their daily rhythm and without discomfort by following the recommendation content according to the recommendation signal, without having to search for an operating time period for additional function operation themselves.

If the air-conditioned space is, for example, a living room where the user spends most of their time outside of sleeping hours, the air conditioner 2 predicts the timing when the user changes from absent to present in the 24-hour presence/absence prediction results for the air-conditioned space as the user's wake-up time or return home time. Then, the air conditioner 2 transmits a recommendation signal to the user's terminal device 7 recommending the setting of the start time of air-conditioning operation, based on the timing when the user changes from absent to present in the living room, for example. As a result, the user can easily set the start time of air-conditioning operation in the living room at the time of wake-up time.

The air conditioner 2 predicts the timing when the user changes from being present to being absent in the 24-hour presence/absence prediction results for the air-conditioned space corresponding to the indoor use information as the time when the user leaves or goes to bed. The air conditioner 2 then transmits a recommendation signal to the user's terminal device 7, recommending additional function operation other than air conditioning operation, based on the timing when the user changes from being present to being absent in the living room. As a result, the user can easily set additional function operation in the living room when the user leaves or goes to bed, without having to search for the operating time period for the additional function operation themselves.

When the air-conditioned space is, for example, a room used by a user when going to sleep, such as a bedroom, the air conditioner 2 predicts the timing when the user changes from absent to present in the 24-hour presence/absence prediction results for the air-conditioned space as the user's bedtime. The air conditioner 2 then transmits a recommendation signal to the user's terminal device 7, recommending the setting of the start time of air-conditioning operation, based on the timing when the user changes from absent to present in the bedroom, for example. As a result, the user can easily set the start time of air-conditioning operation in the bedroom at bedtime.

The air conditioner 2 predicts the timing when the user changes from being present to being absent in the 24-hour presence/absence prediction results for the air-conditioned space corresponding to the indoor use information as the user's wake-up time. The air conditioner 2 then transmits a recommendation signal to the user's terminal device 7, recommending additional function operation other than air conditioning operation, based on the timing when the user changes from being present to being absent in the bedroom. As a result, the user can easily set additional function operation in the bedroom at the wake-up time without having to search for the operating time period for the additional function operation themselves.

The air conditioner 2 predicts the time when the user wakes up or returns home based on the 24-hour presence/absence prediction results for each air-conditioned space corresponding to the indoor use information, when the predicted absence period changes from a fourth predetermined period, for example, 3 hours or more, to presence. As a result, the air conditioner 2 can predict the user's wake-up time or return home time with high accuracy.

Based on the 24-hour presence/absence prediction results for each air-conditioned space corresponding to the indoor use information, the air conditioner 2 predicts the start of a predicted absence period of a fifth predetermined period, for example, an absence that continues for three hours or more, as the time the user leaves or goes to bed. As a result, the air conditioner 2 can predict the time the user leaves or goes to bed with high accuracy.

The air conditioner 2 transmits a recommendation signal to the user's terminal device 7 recommending additional function operation other than air conditioning operation when the predicted period of absence is a sixth predetermined period, for example, 5 hours or more, based on the 24-hour presence/absence prediction results for each air-conditioned space corresponding to the indoor use information. As a result, the user can easily set additional function operation at a time period where the user will not feel uncomfortable, without having to search for a time period for additional function operation themselves.

<Modification of the First Embodiment>
In the communication adapter 3 of the first embodiment, a presence/absence pattern to be used for prediction is selected from among a plurality of presence/absence patterns using the presence/absence detection result, day of the week information, and holiday information, and the selected presence/absence pattern is used to predict the presence/absence of the user in the air-conditioned space, but the server device 5 may predict the presence/absence of the user in the air-conditioned space. In this case, the server device 5 selects a presence/absence pattern to be used for prediction from a predetermined time to a predetermined time before the time when the presence/absence of the user in the air-conditioned space is predicted, the day of the week information, and holiday information from among a plurality of presence/absence patterns, and predicts the presence/absence of the user in the air-conditioned space using the selected presence/absence pattern. Then, the server device 5 transmits the presence/absence prediction result to the air conditioner 2 via the communication adapter 3. As a result, the server device 5 can generate a presence/absence pattern and predict the presence/absence, thereby reducing the processing load on the communication adapter 3 side.

In the air conditioning system 1 of Example 1, the indoor unit 21 of the air conditioner 2 transmits the recommendation signal to the user's terminal device 7 via the communication adapter 3, but the signal may be transmitted to the remote control 23 of the air conditioner 2, and this can be modified as appropriate.

The present example shows a case in which the presence/absence prediction unit 34E selects a presence/absence pattern to be used for prediction from among multiple presence/absence patterns using the presence/absence detection results from a specified time to a specified time ago, day of the week information, and holiday information, and predicts presence/absence for a 24-hour period in the air-conditioned space using the selected presence/absence pattern. However, even if there is no holiday information, the prediction unit 34E may select a presence/absence pattern to be used for prediction using the presence/absence detection results from a specified time to a specified time ago and day of the week information.

The recommendation unit 21D0 selects a recommended operation to be recommended to the user based on the user's behavior prediction result (time to return home, time to go out, time to bed, etc.) predicted by the behavior prediction unit 21D2, and recommends the execution of the selected recommended operation. However, the recommendation unit 21D0 may select a recommended operation to be recommended to the user based on the user's behavior prediction result, and execute the selected recommended operation. As a result, the user can be saved the trouble of checking the recommended operation to be recommended.

In addition, the air conditioning system 1 uses the air conditioner 2, the communication adapter 3, and the server device 5, and the generation of the presence/absence pattern is performed by the server device 5, the prediction of presence/absence is performed by the communication adapter 3, and the transmission of the signal recommending the start of air conditioning operation is performed by the indoor unit 21 of the air conditioner 2. However, the prediction of presence/absence and the transmission of the signal recommending the start of air conditioning operation may be performed by the communication adapter 3, that is, all of the processes in Figures 13B, 13C, and 13D may be performed by the communication adapter 3, and modifications may be made as appropriate.

For ease of explanation, a user who works during the day has been exemplified, and, for example, when there is a timing during the first time period when the user changes from being absent for three or more consecutive hours to being present, the timing is predicted as the time to wake up. However, in the case of a user who works at night, the timing may be predicted as the time to return home, and can be changed as appropriate.

The air conditioner 2 may also be configured to generate a presence/absence pattern, predict presence/absence, and generate and transmit a signal recommending the start of air conditioning operation. This embodiment is described below as Example 2. Note that the same components as in Example 1 are given the same reference numerals, and descriptions of the overlapping components and operations are omitted.

<Air conditioner configuration>
Fig. 14 is a block diagram showing an example of the configuration of an air conditioner 2A of Example 2. In addition to the main body 21A, the sensor 21B, the light receiving unit 21C, and the control unit 21D, the indoor unit 210 in the air conditioner 2A shown in Fig. 14 has an acquisition unit 21E, a presence/absence pattern 21F, a generation unit 21G, and a presence/absence prediction unit 21H. The acquisition unit 21E acquires day of the week information. The presence/absence pattern 21F is a pattern generated for each day of the week as a presence/absence pattern indicating the presence/absence of a user in the air-conditioned space.

The generation unit 21G generates the presence/absence pattern 21F using the presence/absence detection result of the sensor 21B, the day of the week information, and the holiday information. When a holiday is included in the time period of the presence/absence detection result of the sensor 21B, the generation unit 21G considers the time period to be the same as a holiday. Among the presence/absence detection results, the presence/absence detection result of "indefinite" is not used for the presence/absence pattern 21F. The presence/absence prediction unit 21H selects a presence/absence pattern to be used for prediction from among the multiple presence/absence patterns 21F, using the presence/absence detection result from a specified time, which is the time when the presence/absence of the user is predicted, to a specified time before. Furthermore, the presence/absence prediction unit 21H predicts the presence/absence of the user in the air-conditioned space using the selected presence/absence pattern. The behavior prediction unit 21D2 in the control unit 21D transmits a recommendation signal recommending various air conditioning operations to the user's terminal device 7 based on the presence/absence prediction result of the user in the air-conditioned space by the presence/absence prediction unit 21H.

The generation unit 21G generates the presence/absence pattern 21F using the presence/absence detection results detected by the sensor 21B over a first predetermined period, for example, 30 days. The generation unit 21G stores the presence/absence detection results of the sensor 21B in a storage unit (not shown) without going through the communication adapter 3, and executes the generation process shown in FIG. 12 to generate or update the presence/absence pattern 21F using the stored presence/absence detection results.

The presence/absence prediction unit 21H uses the selected presence/absence pattern to predict the presence/absence of a user in the air-conditioned space at a predetermined time, for example, 8:00 or 20:00 every day, for a second predetermined period from the predetermined time, for example, up to 24 hours later. The presence/absence prediction unit 21H predicts the presence/absence of a user in the air-conditioned space for a third predetermined period, for example, every 10 minutes. If a predicted time period includes a public holiday, the presence/absence prediction unit 21H treats the time period as the same as a public holiday and predicts the presence/absence in the air-conditioned space for 24 hours. In addition, the presence/absence prediction unit 21H excludes the presence/absence detection result of "indeterminate" from the presence/absence detection result of the sensor 21B used when predicting the presence/absence of a user in the air-conditioned space. The presence/absence prediction unit 21H executes the processes of steps S30, S31, S32, S33, and S34 shown in FIG. 13A.

The behavior prediction unit 21D2 predicts the user's daily behavior in the living space (e.g., wake-up time, time to go out, time to return home, time to go to bed) based on the prediction result of the presence/absence prediction unit 21H of the user's presence/absence in the air-conditioned space and the indoor use information of the air-conditioned space (living room or bedroom). The recommendation unit 21D0 then selects an operation to be recommended to the user based on the predicted behavior. Furthermore, the recommendation unit 21D0 recommends to the user to perform the operation to be recommended to the user. The recommendation unit 21D0 transmits a recommendation signal to the remote control 23, which recommends to the user. Furthermore, the recommendation unit 21D0 executes the air conditioning operation recommendation process in the living room in step S35 shown in FIG. 13A, or the air conditioning operation recommendation process in the bedroom in step S36 shown in FIG. 13A.

The extraction unit 21D1 extracts time periods in which additional function operations different from normal air conditioning operations can be performed from the time periods of the 24-hour presence/absence prediction results in the air-conditioned space. The time periods in which additional function operations can be performed are, for example, absence periods of 5 hours or more. The recommendation unit 21D0 transmits a recommendation signal to the remote control 23 recommending additional function operations such as heating sterilization operation and filter cleaning operation different from air conditioning operations based on the presence/absence prediction results of the user in the air-conditioned space by the presence/absence prediction unit 21H. Note that, although an example has been given of the case in which the recommendation unit 21D0 transmits a recommendation signal to the remote control 23, if the recommendation unit 21D0 has a means of communicating with the user's terminal device 7, it may transmit a recommendation signal to the user's terminal device 7, and this can be changed as appropriate.

Effects of Example 2
The air conditioner 2A of the second embodiment selects one presence/absence pattern from among a plurality of presence/absence patterns using at least the current detection result of the sensor 21B. The air conditioner 2A predicts the presence/absence in the air-conditioned space for 24 hours using the selected presence/absence pattern. Furthermore, the air conditioner 2A transmits a recommendation signal to the remote control 23 recommending various air conditioning operations based on the presence/absence prediction results for the 24 hours. As a result, the user can achieve air conditioning operation in accordance with their lifestyle by implementing the recommended content according to the recommendation signal, thereby obtaining a comfortable air-conditioned space.

The air conditioner 2A generates a presence/absence pattern using the presence/absence detection results, day of the week information, and holiday information detected by the sensor 21B over a first predetermined period, for example 30 days. As a result, a highly accurate presence/absence pattern in the air-conditioned space can be obtained.

The air conditioner 2A uses the selected presence/absence pattern to predict, at a specified time, for example 8:00 or 20:00 every day, a 24-hour presence/absence prediction that indicates the presence or absence of a user in the air-conditioned space for a second specified period, for example up to 24 hours, from the specified time.The air conditioner 2A then transmits a recommendation signal to the remote control 23 that recommends various air conditioning operations based on the 24-hour presence/absence prediction results.As a result, the user can receive recommendations for air conditioning operations that are in line with their lifestyle rhythm, using the presence/absence prediction results for the period up to 24 hours from the specified time.

The air conditioner 2A uses the selected presence/absence pattern to predict the presence/absence of a user in the air-conditioned space for a third predetermined period, for example, every 10 minutes for 24 hours.Then, the air conditioner 2A transmits a recommendation signal to the remote control 23 recommending various air conditioning operations based on the presence/absence prediction results for the 24 hours.As a result, the user can receive recommendations for air conditioning operations that are in line with their lifestyle rhythm using the presence/absence prediction results for every 10 minutes for the 24 hours.

The air conditioner 2A transmits a recommendation signal to the remote control 23 recommending the setting of the air conditioning operation start time or the air conditioning operation stop time based on the predicted 24-hour presence/absence prediction result in the air-conditioned space. As a result, the user can realize a comfortable air-conditioned space that matches the rhythm of their life by following the recommendation content according to the recommendation signal, without having to search for an air-conditioning operation time period such as sleep operation.

The air conditioner 2A transmits a recommendation signal to the remote control 23 recommending additional function operation other than air conditioning operation based on the predicted 24-hour presence/absence prediction results in the air-conditioned space. As a result, the user can realize additional function operation in accordance with their lifestyle rhythm and without discomfort by following the recommendation content according to the recommendation signal, without having to search for an operating time period for additional function operation themselves.

When the air-conditioned space is, for example, a living room where the user spends most of their time outside of bedtime, the behavior prediction unit 21D2 predicts the timing when the user changes from absent to present in the air-conditioned space based on the 24-hour presence/absence prediction results for the air-conditioned space as the user's wake-up time or return home time. Furthermore, the recommendation unit 21D0 transmits a recommendation signal to the remote control 23 recommending the setting of the start time of the air-conditioning operation, for example, based on the timing when the user changes from absent to present in the living room. As a result, the user can easily set the start time of the air-conditioning operation in the living room at the time of wake-up time.

The behavior prediction unit 21D2 predicts the timing when the user changes from being present to being absent in the 24-hour presence/absence prediction results in the air-conditioned space corresponding to the indoor use information as the time when the user leaves or goes to bed. The recommendation unit 21D0 transmits a recommendation signal to the remote control 23 recommending an additional function operation different from air conditioning operation based on the timing when the user changes from being present to being absent in the living room. As a result, the user can easily set additional function operation in the living room at the time when the user leaves or goes to bed without having to search for the operating time period for the additional function operation themselves.

When the air-conditioned space is, for example, a room used by a user when going to sleep, such as a bedroom, the behavior prediction unit 21D2 predicts the timing when the user changes from absent to present in the 24-hour presence/absence prediction result in the air-conditioned space as the user's bedtime. The recommendation unit 21D0 transmits a recommendation signal to the remote control 23 recommending the setting of the start time of air-conditioning operation, for example, based on the timing when the user changes from absent to present in the bedroom. As a result, the user can easily set the start time of air-conditioning operation in the bedroom at bedtime.

The behavior prediction unit 21D2 predicts the timing when the user changes from being present to being absent in the 24-hour presence/absence prediction result in the air-conditioned space corresponding to the indoor use information as the user's wake-up time. The recommendation unit 21D0 transmits a recommendation signal to the remote control 23 recommending additional function operation other than air conditioning operation based on the timing when the user changes from being present to being absent in the bedroom. As a result, the user can easily set additional function operation in the bedroom at the wake-up time without having to search for the operating time period for the additional function operation themselves.

The behavior prediction unit 21D2 predicts the time when the user wakes up or returns home based on the 24-hour presence/absence prediction results for each air-conditioned space corresponding to the indoor use information, when the predicted absence period changes from a fourth predetermined period, for example, 3 hours or more, to presence. As a result, the air conditioner 2A can predict the user's wake-up time or return home time with high accuracy.

The behavior prediction unit 21D2 predicts the start of a fifth predetermined period of predicted absence, for example, an absence lasting three hours or more, based on the 24-hour presence/absence prediction results for each air-conditioned space corresponding to the indoor use information, as the time the user leaves or goes to bed. As a result, the air conditioner 2 can predict the time the user leaves or goes to bed with high accuracy.

The recommendation unit 21D0 transmits a recommendation signal to the remote control 23 that recommends additional function operation other than air conditioning operation when the predicted period of absence is a sixth predetermined period, for example, 5 hours or more, based on the 24-hour presence/absence prediction result for each air-conditioned space corresponding to the indoor use information. As a result, the user can easily set additional function operation without having to search for a time period for additional function operation by himself/herself.

<Modifications of the embodiment>
The first, second, third, fourth, fifth, sixth, and seventh predetermined periods in the first and second embodiments can be changed as appropriate. The first, second, third, and fourth time periods can be changed as appropriate.

In addition, the heating sterilization operation is exemplified as starting a heating sterilization operation for one hour after a monitoring operation, but the operation time is not limited to this and can be changed as appropriate. In addition, the heating sterilization operation is exemplified as ending after a cooling period of three hours after a heating sterilization operation for one hour, but the cooling period can be shortened by performing an air conditioning operation during the cooling period, and can be changed as appropriate.

Also, in the above example, the presence/absence patterns used to predict the user's presence/absence are limited to a maximum of five patterns, taking into consideration the increase in communication volume between the communication adapter 3 and the server device 5 and the increase in memory capacity required by the communication adapter. However, if there is no need to consider the communication volume and the memory capacity can be increased, a presence/absence pattern may be generated for each day of the week, and the presence/absence of the user may be predicted using the presence/absence pattern corresponding to the day of the week to be predicted.

Furthermore, each component of each part shown in the figure does not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each part is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

Furthermore, the various processing functions performed by each device may be executed in whole or in part on a CPU (Central Processing Unit) (or a microcomputer such as an MPU (Micro Processing Unit) or MCU (Micro Controller Unit)). Needless to say, the various processing functions may be executed in whole or in part on a program that is analyzed and executed by a CPU (or a microcomputer such as an MPU or MCU), or on hardware using wired logic.

REFERENCE SIGNS LIST 1 Air conditioning system 2, 2A Air conditioner 3 Communication adapter 5 Server device 7 Terminal device 21, 210 Indoor unit 21B Sensor 21D0 Recommendation unit 21D2 Behavior prediction unit 21E Acquisition unit 21F Presence/absence pattern 21G Generation unit 21H Presence/absence prediction unit 34E Presence/absence prediction unit 53B Acquisition unit 53C Generation unit

Claims (22)

A sensor for detecting the presence or absence of people in the air-conditioned space;
a storage unit that stores a presence /absence prediction result, which is a result of predicting the presence/absence of a user in the air-conditioned space using a presence/absence pattern that indicates a tendency of the user to be present or absent in the air-conditioned space, in association with indoor use information that identifies an indoor use of the air-conditioned space;
a behavior prediction unit that predicts the behavior of the user for a predetermined period of time based on the presence/absence prediction result, the indoor use information of the air-conditioned space, and a time period when air-conditioning operation is expected to be started in the air-conditioned space or a time period when air-conditioning operation is expected to be stopped in the air-conditioned space;
a recommendation unit that selects an operation to be recommended to the user based on a behavior prediction result of the behavior prediction unit;
An air conditioner comprising: The recommendation unit is
2. The air conditioner according to claim 1 , wherein the execution of the selected operation is recommended to the user. The behavior prediction unit is
The air conditioner according to claim 1 or 2, characterized in that, when the indoor use information of the air-conditioned space is a living room, in the prediction result of the presence or absence of the user in the air-conditioned space, the timing when the user changes from absent to present is predicted to be the user's wake-up time or return home time. The behavior prediction unit is
The air conditioner according to claim 3 , characterized in that, when the period of absence of the user in the air-conditioned space is longer than a predetermined length, the timing at which the user will change to presence is predicted to be the user's wake-up time or return home time. The behavior prediction unit is
5. The air conditioner according to claim 3, wherein when the period of absence is equal to or longer than a predetermined length during a first time period, the timing of the change to presence is predicted to be the time when the user gets up. The behavior prediction unit is
5. The air conditioner according to claim 3, wherein when the period of absence is equal to or longer than a predetermined length during the second time period, the timing of the change to presence is predicted to be the time when the user returns home. a recommendation unit that selects an operation to be recommended to the user based on a behavior prediction result of the behavior prediction unit;
The recommendation unit is
The air conditioner according to any one of claims 3 to 6, characterized in that when the wake-up time or the return home time of the user is predicted, air conditioning operation is selected as the operation to be recommended to the user. a recommendation unit that selects an operation to be recommended to the user based on a behavior prediction result of the behavior prediction unit;
The behavior prediction unit is
The air conditioner according to claim 1 or 2, characterized in that, when the indoor use information of the air-conditioned space is a living room, in the prediction result of the presence or absence of the user in the air-conditioned space, the timing when the user changes from being present to being absent is predicted to be the time when the user leaves the room or the time when the user goes to bed. The behavior prediction unit is
The air conditioner according to claim 8, characterized in that if the period of absence of the user in the air-conditioned space is longer than a predetermined length, the timing at which the absence will change to absence is predicted to be the time when the user leaves the house or goes to bed. The behavior prediction unit is
10. The air conditioner according to claim 8 or 9, wherein, if the period of absence is equal to or longer than a predetermined length during a third time period, the timing at which the absence will change to the absence is predicted to be the time at which the user leaves the home. The behavior prediction unit is
10. The air conditioner according to claim 8 or 9, wherein, if the period of absence is equal to or longer than a predetermined length during a fourth time period, the timing at which the absence will change to absence is predicted to be the user's bedtime. a recommendation unit that selects an operation to be recommended to the user based on a behavior prediction result of the behavior prediction unit;
The recommendation unit is
An air conditioner as described in any one of claims 8 to 11, characterized in that when the user's time of leaving is predicted, an additional function operation other than air conditioning operation is selected as the operation to be recommended to the user. The behavior prediction unit is
The air conditioner according to claim 1 or 2, characterized in that, when the indoor use information of the air-conditioned space is a bedroom, in the prediction result of the presence or absence of the user in the air-conditioned space, the timing when the user changes from absent to present is predicted to be the user's bedtime. The behavior prediction unit is
The air conditioner according to claim 13 , characterized in that, when the period of absence of the user in the air-conditioned space is equal to or longer than a predetermined length, the timing at which the user will change to be present is predicted to be the time when the user goes to bed. The behavior prediction unit is
15. The air conditioner according to claim 13, wherein, in a fourth time slot, if the period of absence is equal to or longer than a predetermined length, the timing at which the period of presence will change to be the user's bedtime is predicted. a recommendation unit that selects an operation to be recommended to the user based on a behavior prediction result of the behavior prediction unit;
The recommendation unit is
The air conditioner according to any one of claims 13 to 15 , characterized in that, when the bedtime of the user is predicted, air conditioning operation is selected as the operation to be recommended to the user. The behavior prediction unit is
The air conditioner according to claim 1 or 2, characterized in that, when the indoor use information of the air-conditioned space is a bedroom, in the prediction result of the presence or absence of the user in the air-conditioned space , the timing when the user changes from being present to being absent is predicted to be the time when the user leaves the room. The behavior prediction unit is
The air conditioner according to claim 17 , characterized in that, when the period of absence of the user in the air-conditioned space is equal to or longer than a predetermined length, the timing of the change to absence is predicted to be the time when the user leaves the room. The behavior prediction unit is
The air conditioner according to claim 17 or 18, characterized in that if the period of absence is equal to or longer than a predetermined length during a third time period, the timing at which the absence will change to the absence is predicted to be the time at which the user leaves the room. a recommendation unit that selects an operation to be recommended to the user based on a behavior prediction result of the behavior prediction unit;
The recommendation unit is
An air conditioner as described in any one of claims 17 to 19, characterized in that when the user's departure time is predicted, an additional function operation other than air conditioning operation is selected as the operation to be recommended to the user. The air conditioner further includes an indoor unit having a heat exchanger,
The additional function operation includes:
The air conditioner according to claim 12 , further comprising a heat sterilization operation for sterilizing the heat exchanger by heating it. The sensor includes:
An air conditioner as described in any one of claims 1 to 7, characterized in that the operation of detecting the presence or absence of a person is started in response to a supply of power to the air conditioner, and the operation of detecting the presence or absence of a person is continued regardless of whether air conditioning operation is started or stopped.

Images