JP6308466B2 - Environmental control device, program - Google Patents

Description

  The present invention relates to an environment control apparatus that instructs an operation of a device that controls an environment, and a program that causes a computer to function as the environment control apparatus.

  Conventionally, it has been proposed to control devices that control the environment, such as lighting devices and air conditioning devices, in accordance with user preferences (see, for example, Patent Document 1). In the technology described in Patent Document 1, the environment information sensed by a plurality of sensor nodes is used, and the signal generation unit generates an apparatus control signal for controlling the apparatus using an evaluation function. In order to adjust the environment such as brightness and temperature to the preference of the person, the sensing signal included in the evaluation function is weighted according to the preference of the user. User coefficients for performing weighting for each user are set in the user table. Further, Patent Document 1 describes that the user coefficient set in the user table is changed by an operation signal when the device is operated using a remote controller or the like.

Japanese Patent No. 4602914

  As described above, in the technique described in Patent Document 1, the user coefficient set in the user table in order to reflect the user's preference in the control of the device is changed by an operation of the remote controller or the like. Therefore, the user coefficient is set depending on the state of the user at the time when the remote controller or the like is operated, and it cannot be said that the user coefficient necessarily corresponds to the user's preference. That is, the user coefficient may be changed depending on the mood rather than the user's preference.

  An object of this invention is to provide the environment control apparatus which enabled it to implement | achieve control of the apparatus according to a user preference better. Furthermore, an object of the present invention is to provide a program that causes a computer to function as the environment control apparatus.

  In the environment control device according to the present invention, a storage unit that stores a control pattern that is a change in control amount with the passage of time in a cluster in which sensibility features for each user are classified, and an evaluation for the environment are input. An input unit, a plan unit that generates a control plan using the control pattern extracted from the storage unit by using a sensuous feature based on information about evaluation given to the input unit, and the plan unit generated And an execution unit that controls devices that control the environment according to the control plan.

  The program according to the present invention causes a computer to function as an environment control device.

  According to the structure of this invention, it becomes possible to implement | achieve control of the apparatus according to a user preference better.

It is a block diagram which shows embodiment. It is a figure which shows the concept of the cluster in embodiment. It is a figure which shows the example of a control pattern in embodiment.

  The embodiment described below assumes an environment in which control by a device is possible, and it takes a relatively long time to reach a target state after the start of control of the device. However, the environment to be controlled includes not only physical elements but also elements based on the user's experience. Therefore, the elements of the physical environment are, for example, temperature, humidity, and the like, but may include elements that affect the bodily sensation such as wind speed and direction. Moreover, the environment in a building is assumed, and the explanation will be made assuming an indoor environment in a house.

  A typical example of the device is an air-conditioning device capable of cooling and heating, but the same technology can be applied to a heater, a floor heating device, a dehumidifier, a humidifier, and the like.

  By the way, when the indoor temperature is controlled, the set temperature is changed according to the perception of the user, such that the set temperature is lowered when it gets hot and the set temperature is raised when it gets cold. That is, physical elements in the environment are changed by the set temperature. On the other hand, the user changes the set temperature according to the sensitivity such as “hot”, “cold”, “comfortable”, and the like. Therefore, when the environment is controlled by the device, it is desirable to control the device so that the user feels comfortable rather than information such as temperature and humidity. However, it is difficult to control the device so that the user feels comfortable according to the sense perceived by the user.

  The embodiment described below is intended to make it possible to reflect the user's feeling in the control of the device, thereby improving the user's satisfaction with the environment.

(Embodiment 1)
In this embodiment, the device to be controlled is an air conditioner, and there are four types of environmental elements that can be controlled by the air conditioner: temperature, humidity, wind speed, and wind direction. The space where the air conditioner controls the environment is indoors. The air conditioner of this embodiment assumes a case where the temperature can be set in increments of 0.5 to 1 ° C. The wind speed can be set in a plurality of stages of about 3 to 5 stages, and the wind direction can be adjusted in an angle range of about 50 to 90 degrees in the vertical direction and in an angle range of about 90 to 180 degrees in the left-right direction. The adjustable angle range in the wind direction is not the absolute value of the angle with respect to the floor surface or the like, but means the difference between the maximum angle and the minimum angle that can be adjusted in each direction, that is, the swing width of the angle.

  Further, the humidity is relative humidity, and the air conditioner of the present embodiment has functions of dehumidification and humidification, and the humidity can be set in increments of 5 to 10%. However, it is not essential that the humidity can be set.The humidity can be adjusted if the air-conditioning equipment has a dehumidifying or humidifying function, and even if the relative humidity is changed by changing the temperature, the humidity is Considered adjustable.

  As described above, the air conditioner used in the present embodiment can set the temperature and humidity, and the wind speed and direction can be adjusted. In the following, a configuration example in which temperature is mainly controlled as an environmental element among temperature, humidity, wind speed, and wind direction will be described, but humidity, wind speed, and wind direction are also controlled as necessary.

(overall structure)
As shown in FIG. 1, the environmental control apparatus 10 uses environmental information D11 obtained by the sensor 20 measuring environmental elements as input information, and uses control information D12 that instructs the operation of the air conditioner 30 as output information. . That is, the environment control device 10 includes an input unit 11 to which the environment information D11 is input from the sensor 20, and an output unit 12 that outputs the control information D12 to the air conditioner 30. Environmental information D11 is input from the sensor 20 to the input unit 11 at a predetermined capture cycle, and the output unit 12 outputs control information D12 to the air conditioner 30 at a predetermined control cycle.

  The capture cycle and the control cycle are appropriately determined in the range of about 30 seconds to 10 minutes, for example. The time from when the set temperature of the air conditioner 30 is changed until the room temperature reaches the set temperature (target environment) (hereinafter referred to as “control time”) changes according to the environment. The control time requires, for example, 5 minutes to 1 hour. Therefore, both the capture period and the control period may be set to 1 minute, and the control time from the start of control may be set to 5 minutes. This value is only an example, and can be changed as appropriate.

  The environmental information D11 is transferred from the input unit 11 to the planning unit 13, and the planning unit 13 generates a control plan for the air conditioner 30 using the environmental information D11 and sensitivity information described later. This control plan represents a change in setting contents with the passage of time related to a control target such as temperature, humidity, wind speed, and wind direction for the air conditioner 30. For example, the control plan related to the set temperature of the air conditioner 30 means a change in the set temperature with time.

  A plurality of types of templates are prepared in advance in the control plan, and any one of the templates is adopted according to the sensitivity information. Therefore, the planning unit 13 generates a control plan corresponding to the sensitivity information by selecting a template corresponding to the sensitivity information. The template is set as a typical example of a change in temperature over time when the environmental element controlled by the air conditioner 30 is temperature. Since the set temperature of the air conditioner 30 is reflected in the power consumption, the template may be set as a typical example of a change in power consumption over time.

  The control plan generated by the planning unit 13 is input to the execution unit 14, and the execution unit 14 outputs the set temperature for each control cycle as a control amount according to the elapsed time, and outputs the control amount for each control cycle through the output unit 12. The air conditioner 30 is given.

  By the way, the template described above is stored in the storage unit 15 in association with the sensitivity information. That is, in order for the planning unit 13 to generate a control plan based on the environmental information D11 and the sensitivity information acquired from the sensor 20 and to provide the control information D12 to the air conditioner 30, a template is registered in the storage unit 15 in advance. There must be. In other words, the operation of the environment control device 10 includes a pre-process for creating a template and registering it in the storage unit 15 and a normal process for operating the air conditioner 30 based on the environment monitored by the sensor 20.

(Pre-processing)
In the pre-processing, the input unit 11 receives an answer to a question to the user in order to analyze the user's sensibility characteristics with respect to environmental changes. An answer to the question is input to the input unit 11 from the operation device 41 operated by the user. Although the question can be presented to the user using paper or the like on which the question is printed, in the present embodiment, the operation display unit 40 including the operation unit 41 and the display unit 42 is used to display the question. The structure which presents a question to the device 42 is adopted. This type of operation indicator 40 may be realized by a general-purpose device such as a smartphone, a tablet terminal, or a personal computer in addition to a dedicated device for the environment control device 10.

  In order to grasp the user's characteristics from various aspects, it is desirable that the questions are asked about a plurality of items. The question is set as appropriate, such as content related to the air conditioner 30 and content that evaluates the entire living environment. For example, the contents of the question are set in a form in which an appropriate situation is presented to the user and an ideal environment that the user recalls is answered in response to the presented situation.

  As an example, an item “Imagine an ideal air conditioner cooling environment for you and evaluate it with the following items” is set. The form of this example is not a question form, but is substantially equivalent to a question asking about an ideal environment for the user in the situation of an air conditioner cooling environment.

  For a question, a plurality of types of adjectives or adjective verbs representing a sense of the environment are prepared as answer options, and the user's sensation is selected from these options. For example, for the above-described questions, modifiers representing a thermal environment such as “warm”, “cool”, “fresh”, and “sultry” are set as options. The options are set in a plurality of stages of about 3 to 7 or in the form of “Yes” and “No”. In any case, a format that allows the user to easily answer is used.

  Numerical values are associated with the answer options, and when there are three types of options, for example, an integer value of 1 to 3 is associated with each option. Further, information for identifying the user (user ID) is given to the user who answers the question, and the answer for each question is recorded in the storage unit 15 in association with the user ID and the questioned item.

  In the above-described example, the user is reminded of an appropriate situation, and an answer is requested for the reminded situation. However, it is desirable to include an evaluation of the environment at the time of questioning in the answer. That is, the evaluation of the environment at the time of questioning and the target environment for the recalled situation are stored in the storage unit 15. The evaluation of the environment at the time of the question can be given to the input unit 11 at an arbitrary time, and the user's evaluation of the actual environment can be accumulated in the storage unit 15 by this processing.

  However, it is not essential to give the user's evaluation of the real environment to the storage unit 15, by recalling an appropriate situation and letting the user answer the ideal environment that the user wants, without performing experiments or measurements, You can know how the user feels about the environment. When this type of information is accumulated in the storage unit 15, it is possible to analyze features relating to the user's sense using the information accumulated in the storage unit 15. Analysis related to the user's feeling is performed by the analysis unit 16, and a template based on the analysis result by the analysis unit 16 is stored in the storage unit 15.

  The storage unit 15 described above is preferably provided in a server (not shown) that communicates with the planning unit 13 through an electric communication line (not shown) such as the Internet. The server may be composed of a single computer or a plurality of computers. Further, the server may be configured with a cloud computing system. When the storage unit 15 is provided in the server, the analysis unit 16 is also preferably provided in the server. In this configuration, the planning unit 13 delivers the environment information D11 and the sensitivity information to the storage unit 15 through the telecommunication line, and receives the template from the storage unit 15 through the telecommunication line.

  The configuration example described above uses the environment control apparatus 10 not only for normal processing but also for preprocessing. On the other hand, it is also possible to provide a device that performs preprocessing separately from the environment control device 10 that performs normal processing. For example, the clustering described below can be performed by a device different from the environment control device 10, and only the clustering result can be stored in the storage unit 15.

(Clustering)
The storage unit 15 stores combinations of questions and answers for many users, and the analysis unit 16 performs clustering on the combinations of questions and answers stored in the storage unit 15. When clustering is performed, it is possible to divide a combination of a question and an answer into a plurality of clusters based on the user's sensibility characteristics for the environment. In other words, the combination of question and answer is set to allow classification of the user's emotional characteristics with respect to the environment.

  During clustering, the analysis unit 16 performs cluster analysis in a multidimensional vector space having question items as axes. That is, cluster analysis is performed on a multidimensional vector having the numerical value of the answer for each question as a vector component. When such a multidimensional vector is arranged in the vector space, the vector distribution is biased, and a cluster is formed by vectors that are relatively close to each other. A vector having a short distance can be regarded as a set of users having similar sensibilities.

  Several types of algorithms have been proposed for cluster analysis. Each algorithm has a difference in distance calculation method, cluster formation rule, and the like. The analysis unit 16 performs clustering on a set of a question and an answer by properly using an appropriate algorithm. The cluster is preferably determined automatically by an automatic cutting algorithm, but can be manually determined by an operator. When separating clusters, it is desirable to limit the number of clusters to be generated.

  The analysis unit 16 determines a vector representing the cluster (representative vector) for the cluster obtained by the cluster analysis, and gives identification information (cluster ID) for identifying the cluster. The representative vector is set using an average value or median value for each component of the vectors belonging to the cluster. The cluster representative vector is stored in the storage unit 15 in association with the cluster ID.

  Further, the analysis unit 16 stores the vector for each user in the storage unit 15 in association with the user ID and the cluster ID. The cluster ID corresponding to the user ID is determined based on the distance between the vector for each user and the representative vector. That is, the cluster having the smallest distance between the representative vector and the vector for each user is regarded as the cluster to which the corresponding user belongs, and the cluster ID of this cluster is associated with the user ID of the corresponding user. If the contents of the question and the answer options are set appropriately, the cluster will reflect the user's emotional characteristics. Therefore, by defining the cluster to which the user belongs, the user's emotional characteristics will be classified. Become.

  The distance between the vector for each user and the representative vector of the cluster may be selected from Euclidean distance, Mahalanobis distance, and the like. The Euclidean distance is obtained as the square root of the residual sum of squares for each component of the two vectors for which the distance is obtained, but the distance may be evaluated using the residual sum of squares, or a specific component effective for clustering. If there is, the corresponding component may be weighted.

  When there are a large number of items to be inquired, the vector to be clustered has a large number of components. Therefore, the analysis unit 16 may perform cluster analysis after compressing the original multidimensional vector into a low-dimensional vector by principal component analysis. In this case, since the representative vector of the cluster also has a low dimension, it is necessary to compress the user's vector to a low dimension using the vector axis obtained by the principal component analysis in the normal processing.

  In order to compress the original multidimensional vector into a low-dimensional vector, principal component analysis is performed using a plurality of multidimensional vectors stored in the storage unit 15 and subjected to clustering. Of the plurality of principal components obtained by the principal component analysis, when the first principal component and the second principal component are used, the original multidimensional vector is compressed into a two-dimensional vector, and the first principal component, the second principal component, and the first principal component are compressed. When the three principal components are used, the original multidimensional vector is compressed into a three-dimensional vector.

  For example, when the first principal component and the second principal component are used, the original multidimensional vector is projected onto a two-dimensional vector plane having the first principal component and the second principal component as axes as shown in FIG. The coordinate value at that time is a component of a two-dimensional vector. Similarly, when the first principal component, the second principal component, and the third principal component are used, the coordinate values in the three-dimensional vector space are set as the components of the three-dimensional vector. The principal component analysis may be performed using a known algorithm.

  When the original multidimensional vector is compressed into a two-dimensional vector or a three-dimensional vector by performing principal component analysis, a vector for converting the original multidimensional vector into a two-dimensional vector or a three-dimensional vector is obtained. This vector is called an eigenvector. The eigenvector component includes a coefficient for converting the original multidimensional vector into a two-dimensional vector or a three-dimensional vector. An eigenvector is obtained for each principal component.

Assume that the number of questions is n. In addition, a multidimensional vector whose component is an answer to a question for the user k is expressed in the form (Xk1, Xk2,..., Xkn). The multi-dimensional vector components Xk1, Xk2,... Are expressed numerically as described above. When the above-described multidimensional vector is projected onto a two-dimensional plane having the first principal component and the second principal component extracted by principal component analysis as axes, the coordinate values (Pk1, Pk2) on the two-dimensional plane are as follows: Expressed in the format.
Pk1 = a11 · Xk1 + a12 · Xk2 + ... + a1n · Xkn
Pk2 = a21 * Xk1 + a22 * Xk2 + ... + a2n * Xkn
This formula becomes a conversion formula for converting a multidimensional vector into a two-dimensional vector. Here, the vector (a11, a12,..., A1n) consisting of the coefficient group of the upper conversion formula becomes the eigenvector of the first principal component, and the vector (a21, a22,. a2n) is the eigenvector of the second principal component.

  The analysis unit 16 performs cluster analysis on a two-dimensional vector or a three-dimensional vector obtained by compressing the original multidimensional vector. In this cluster analysis, the number of components to be handled is reduced compared to the original multidimensional vector, so that the processing load required for the cluster analysis is reduced. When clustering is performed using principal component analysis in this way, a two-dimensional vector representing a cluster or a component (coordinate value) of a three-dimensional vector is obtained as a principal component score.

  A two-dimensional vector or a three-dimensional vector generated using principal component analysis can be associated with the original multi-dimensional vector on a one-to-one basis. Accordingly, it is possible to obtain a representative vector using a multidimensional vector belonging to a cluster obtained using a two-dimensional vector or a three-dimensional vector. The representative vector thus obtained is stored in the storage unit 15 together with the principal component score. In FIG. 2, the position of the principal component score corresponding to the representative vector for each of the clusters CL1 to CL6 and CLk is shown on a two-dimensional plane.

  The multidimensional vector obtained for each user is converted into a two-dimensional vector or a three-dimensional vector using an already obtained eigenvector. That is, a multidimensional vector for each user is mapped to a two-dimensional plane or a three-dimensional space. The analysis unit 16 obtains the distance between the two-dimensional vector or the three-dimensional vector obtained for the specific user and the two-dimensional vector or the three-dimensional vector representing the representative point for each cluster. The analysis unit 16 regards the cluster having the smallest obtained distance as the cluster to which the user belongs, and stores the original multidimensional vector related to the user in the storage unit 15 in association with the user ID and the cluster ID. The storage unit 15 may store a two-dimensional vector or a three-dimensional vector converted from a multidimensional vector.

  In addition, when the distance calculated | required in order to determine the cluster to which a user belongs does not have a significant difference with respect to several clusters, the cluster to which a user belongs cannot be defined uniquely. That is, a reference value for determining a difference in distance is set, and when the difference in distance is smaller than the reference value for a plurality of clusters, it is determined that there is no significant difference. In such a case, the user may belong to a plurality of clusters. Also, focusing on the component that maximizes the difference between the clusters in the original multidimensional vector, the cluster to which the user belongs the cluster that minimizes the difference between the corresponding component in the multidimensional vector corresponding to the user and the representative vector of the cluster May be adopted. Alternatively, all the components of the multi-dimensional vector corresponding to the user and the representative vector of the cluster are used, and the cluster in which the average value of the difference for each component is the minimum, or the cluster in which the sum of the differences for each component is the minimum, It may be a cluster to which the user belongs.

  A well-known program prepared as a library may be used as the cluster analysis program and the main factor analysis program.

(Create template)
In the storage unit 15, in addition to the cluster to which the user belongs, a template related to the control plan of the air conditioner 30 is registered for each cluster. The template includes a control pattern. As shown in FIG. 3, the control pattern is set as a pattern of change in the control amount with time (hereinafter referred to as “control pattern”). The control amount corresponds to, for example, input power. FIG. 3A shows a control pattern in which the control amount is a logarithmic function of time, FIG. 3B shows a control pattern in which the control amount is a linear function of time, and FIG. 3C shows a control pattern in which the control amount is an exponential function of time. FIG. 3D shows a control pattern in which the control amount oscillates with time and increases. The control pattern shown in FIG. 3 is an example, and for example, a quadratic function can be used. An example of the template will be described later.

  Now, it is assumed that the control pattern corresponds to a quadratic function in the template extracted by the planning unit 13 from the storage unit 15. Further, it is assumed that the environmental control target is temperature (room temperature), the temperature achieved by the control is 23 ° C., and the temperature θ 1 in the environmental information D 11 captured from the sensor 20 is 19 ° C. In this example, the temperature difference Δθ that is changed from the current room temperature to the target room temperature is 4 ° C. The planning unit 13 uses the information, the control period (for example, 1 minute) of the air conditioner 30 and the control time N (for example, 5 minutes) from the start of the control, for the air conditioner 30 for each control period. Assign control amount. If there are a plurality of types of devices that control the environment, the planning unit 13 assigns a control amount to each device.

In the above example, since a quadratic function is selected as the control pattern, the temperature after t minutes from the start of control is expressed as (Δθ / N ² ) × t ² + θ1 = (4/25) × t2 + 19. Is done. Using this equation, a control plan in 1 minute increments from 1 minute to 5 minutes later is generated as shown in Table 1. In Table 1, it is assumed that the air conditioner 30 is capable of setting the temperature in increments of 0.5 ° C.

また、センサ２０から取り込んだ温度が１８℃であり、目標の温度が２０℃であるときに、制御パターンが一次関数であり、制御開始から１時間で目標の温度に到達させるとすれば、ｔ分後の温度は、（２／６０）×ｔ＋１８と表される。したがって、計画部１３は、表１と同様に１分ごとの制御量を求めて、制御計画を生成する。

Further, when the temperature taken in from the sensor 20 is 18 ° C. and the target temperature is 20 ° C., if the control pattern is a linear function and the target temperature is reached in one hour from the start of control, t The temperature after minutes is expressed as (2/60) × t + 18. Therefore, the planning unit 13 obtains a control amount per minute as in Table 1 and generates a control plan.

  In the above-described example, the control pattern has a constant time interval, but the time interval may be variable. If the time interval is variable, the control pattern can be realized by adjusting the time required to change the temperature by 0.5 ° C.

  Here, it is assumed that the emotional characteristics of each user correspond to the cluster to which the user belongs. Therefore, by selecting an appropriate user to experience a controlled environment and evaluating the environment, the environment evaluated by the corresponding user is extracted, and the control plan that generated this environment is used as a template. Is associated with the cluster to which the user belongs. The environment control plan to be experienced by the user is generated by combining controllable environment elements (temperature, humidity, air volume, wind direction, etc.) and combining each combination with a control pattern.

  The environment that the user evaluates as comfortable depends on the relationship between the environment before control (hereinafter referred to as “initial environment”) and the environment after control (hereinafter referred to as “target environment”). The plan template includes the initial environment and the target environment of the air conditioner 30 as parameters. That is, for each cluster, a control pattern with time and information on the initial environment and the target environment are registered in the storage unit 15 as a control plan template. The information on the initial environment and the target environment is preferably an absolute value such as a temperature that is the initial environment and a temperature that is the target environment, but may be a temperature difference between the temperature of the initial environment and the temperature of the target environment. For example, information such as 1 ° C., 3 ° C., and 5 ° C. may be set in the template as the temperature difference.

  The storage unit 15 may store the user evaluation result together with the template. The user's evaluation result may be an evaluation result in the middle of an environmental change generated by a control plan that is a template. An example template is shown in Table 2.

表２に記載されたテンプレートのうち、「制御」は、図３および表１に示した制御パターンを表しており、「クラスタ」は、ユーザが属するクラスタを表している。なお、「評価」は、空調機器３０を使用するユーザが想定した理想の環境を表す評価（問に対する回答）であり、「満足度」は後述するように、ユーザからのフィードバックである。

Of the templates described in Table 2, “Control” represents the control pattern shown in FIG. 3 and Table 1, and “Cluster” represents the cluster to which the user belongs. Note that “evaluation” is an evaluation (answer to the question) that represents an ideal environment assumed by the user who uses the air conditioner 30, and “satisfaction” is feedback from the user, as will be described later.

(Control plan generation)
When the template is registered in the storage unit 15, the air conditioner 30 can be controlled based on the environment information D <b> 11 captured from the sensor 20. Here, the environment control apparatus 10 presents a question about the environment on the display device 42 as in the case of clustering. When the user makes an answer to the environment using the operation device 41, the planning unit 13 compares the answer set with the storage unit 15 to extract a cluster to which the user belongs, and stores the answer set and the extracted cluster in association with each other. Registered in the section 15.

  The planning unit 13 extracts from the storage unit 15 the template having the highest similarity between the “evaluation” and the answer set in the template. If the user is specified, the planning unit 13 extracts a template associated with the cluster to which the user belongs from the storage unit 15.

  The planning unit 13 compares the set of answers input by the user with the “evaluation” of the template stored in the storage unit 15, and the user belongs to the cluster having the minimum distance between the vectors as described above. Find as a cluster candidate. When the candidate cluster to which the user belongs is determined, the planning unit 13 compares the current environment (temperature, humidity, wind speed, wind direction, etc.) measured by the sensor 20 with the initial environment stored in the storage unit 15.

  The planning unit 13 extracts a template whose initial environment is closest to the current environment measured by the sensor 20 from the templates corresponding to the cluster to which the user belongs. By using the information included in this template, the planning unit 13 sets a control plan for controlling the air conditioner 30.

  By the way, setting the temperature of the air conditioner 30 by the user is nothing but setting a target value for the environment. When the user determines a target value, the planning unit 13 can extract a template by collating with the target environment stored in the storage unit 15. Further, as described above, when the template includes a temperature difference, the temperature difference between the environment measured by the sensor 20 and the target value input by the user is collated with the storage unit 15.

  Here, when the planning unit 13 extracts the template, one of the current environment and the environment set by the user may be adopted. When the environment set by the user is adopted, any one of the elements of the environment is selected. Adopt it. These conditions can be applied regardless of the user, and different conditions can be applied for each user.

  The planning unit 13 generates a control plan using the template extracted from the storage unit 15. When the control of the air conditioner 30 is started, the planning unit 13 delivers the control amount to the execution unit 14 every control cycle, and the execution unit 14 delivers the control information D12 including the control amount to the air conditioner 30 through the output unit 12. That is, the set temperature of the air conditioner 30 is changed with time. In the example described above, the set temperature of the air conditioner 30 is updated in increments of 1 minute, for example. In the operation example described above, only the temperature is described among the environmental control targets. However, when a plurality of targets such as temperature and humidity are controlled, a control plan is generated for each target. However, the control plan set as described above is a candidate, and the control plan is determined by the following processing.

(Determination of control plan)
The user can evaluate the environment through the input unit 11 at an arbitrary point in a state where the air conditioner 30 is operating according to the provisional control plan. That is, when the control of the air conditioner 30 is started, the environment control device 10 displays a screen for accepting evaluation by the user on the operation display 40 and waits for input of evaluation from the operation display 40. The environment is evaluated by allowing the user to input an index such as comfort level. Also, multiple questions are asked in the same way as the questions used to generate the cluster, and evaluation is performed using the distance between the vector represented by the answer set for the question and the vector represented by the target answer set. May be performed.

The final control plan is as follows depending on whether or not the evaluation by the user is input during the period when the air conditioner 30 is operating with the temporary control plan, and the evaluation result when it is input. Generated.
(1) When the evaluation is not input: The air conditioner 30 is controlled according to the control plan, and when the control based on the control plan ends, the environment reached by the control plan is in a final state.
(2) When the evaluation is input and the evaluation is high: The environment at the time when the evaluation is input is set as the final state, and the subsequent control plan is discarded. In this case, a new template is stored in the storage unit 15 based on the control pattern from the start of control to the time point when the evaluation is input.
(3) When the evaluation is input and the evaluation is low: The planning unit 13 newly generates a control plan that satisfies the environment information D11 captured from the sensor 20 and the target environment when the evaluation is input.

  Since the evaluation is expressed by the index or vector distance as described above, an appropriate threshold is set, and the evaluation is determined to be high when the evaluation is equal to or greater than the threshold, and the evaluation is less than the threshold It is determined that the evaluation is low.

  If no rating is entered, the user is considered satisfied with the control plan. In this case, the control plan generated by the planning unit 13 is added to the storage unit 15 with various conditions such as evaluation (evaluation idealed by the user), initial environment, target environment, control time, and evaluation result (satisfaction level). Store. That is, this control plan is adopted. When the evaluation is input and the evaluation is high, the adopted control plan is stored in the storage unit 15 together with various conditions as in the case where the evaluation is not input.

  On the other hand, when the evaluation is input and the evaluation is low, the control plan up to the target environment is regenerated, and the control of the air conditioning equipment 30 is continued. However, if the target environment is realized at this stage, templates having different target environments are selected from the storage unit 15. Further, the control plan may be newly generated not only by changing the template but also by changing the control time, the environment at the start of control, the final state environment, or the cluster.

  In this way, when the evaluation is low, that is, for the control plan in which the user is dissatisfied, if the control plan is changed and finally a high evaluation is obtained for the changed control plan, The control plan is stored in the storage unit 15 together with various conditions. That is, when the control plan is different from the existing template, the changed control plan is stored in the storage unit 15 as a new template.

  Regardless of the evaluation result by the user, the user ID is associated with the control plan stored in the storage unit 15. In other words, a control plan that the user considers comfortable is stored in the storage unit 15 for each user. Therefore, the user ID is added to the control plan evaluated by the user.

  In order to identify the user, the user may make a self-report using the operation display 40. When a plurality of users use the air conditioner 30, personal registration is performed in advance in the environment control device 10, and user options may be presented on the operation display 40 when the air conditioner 30 is operated. Moreover, when each user possesses RFID (Radio Frequency IDentification) for identifying the user, the environment control device 10 may be configured to read the information on the RFID and identify the individual. . Alternatively, the user may be specified by performing face authentication or biometric authentication when entering the room where the air conditioner 30 is arranged.

  In addition, the control plan applicable to an applicable user by implementing a control plan sequentially about the cluster to which the specific user who uses the air conditioner 30 belongs within the range of the cluster corresponding to the some user who uses the air conditioner 30. You may ask for. For example, when used in a house, if a cluster to which an individual family belongs is known, a control plan applicable to the corresponding user can be obtained by sequentially executing the control plan for each cluster.

  In the configuration example described above, clustering is performed only once. However, it is desirable to perform clustering when a new combination of question and answer is obtained. Further, clustering may be performed at regular time intervals. When clustering is performed, it is desirable to correct the representative vector for each cluster as necessary, and the correspondence of the template to the cluster is also corrected as necessary.

  As is clear from the above-described operation, when operating the air conditioner 30, the normal operation of instructing the target environment (for example, set temperature) to the air conditioner 30 and the environment formed by the user by the air conditioner 30 are evaluated. There is movement. Therefore, it is desirable that the environment control device 10 is configured so that the user can select one of the operations.

  In addition, when the user does not instruct which operation to be performed by the air conditioner 30, the environment control apparatus 10 waits for the target environment in the initial setting, and waits for input of the evaluation value after the target environment is input. It may be. In this case, the control plan is stored in the storage unit 15 when the control by the control plan is completed. Thereafter, it is determined whether or not to adopt the control plan generated by the planning unit 13.

  Information such as the initial environment, target environment, control time, and evaluation result is recorded in the storage unit 15 in association with the control pattern every time the air conditioner 30 is actually controlled. What is necessary is just to produce | generate a control plan by calculating | requiring a value. An average value or the like may be adopted as the representative value. Moreover, you may use the value weighted so that a weighting coefficient may become large, so that it is new information rather than an average value. If a control plan is generated in this way, it becomes possible to form an agreement among a plurality of users.

  Further, the user ID, the control plan, the target environment, the evaluation result, and the cluster may be associated with the control pattern and recorded in the storage unit 15, and information may be managed for each user. In this case, it is possible to generate a control plan for each user.

  The environment control apparatus 10 of this embodiment includes an input unit 11, a planning unit 13, an execution unit 14, and a storage unit 15. The input unit 11 receives an evaluation for the environment. The planning unit 13 generates a control plan using the control pattern extracted from the storage unit 15 using the sensuous features based on the information regarding the evaluation given to the input unit 11. The execution unit 14 controls the device (air conditioner 30) that controls the environment according to the control plan generated by the planning unit 13. The storage unit 15 stores a control pattern, which is a change in the control amount with time, in association with a cluster in which the sensuous features for each user are classified.

  According to this configuration, an environment in which the user feels comfortable is posted based on the user's emotional characteristics without instructing specific values related to the environment by performing an evaluation according to the sensitivity to the environment. . If the cluster is subdivided using user attributes (age, gender, etc.), it becomes possible to classify the user's emotional features more accurately.

  Moreover, it is desirable that the input unit 11 is configured to receive an evaluation on the environment during a period in which the device (air conditioning device 30) is controlled according to the control plan. In addition, the planning unit 13 is preferably configured to modify or change the control plan when the evaluation represents dissatisfaction.

  According to this configuration, if the user is dissatisfied with the environment generated by the air conditioner 30 according to the control plan, the evaluation is fed back and the control plan is corrected or changed, so that a control plan that satisfies the user is obtained. It becomes like this.

  The storage unit 15 is preferably configured to preferentially accumulate control patterns whose user evaluation is greater than or equal to a reference value.

  According to this configuration, since the storage unit 15 preferentially accumulates the control patterns preferred by the user, it is possible to create an environment that is highly satisfactory for the user.

  The environment control device 10 described above is configured with a processor that operates according to a program and an interface circuit for connecting other devices as main hardware elements. In addition to the configuration that requires a separate memory, this type of processor can use a microcomputer integrated with the memory. That is, by executing an appropriate program on a computer, the computer can function as the environment control apparatus 10 described above.

  The program is previously written in a ROM (Read Only Memory) or provided by a computer-readable recording medium. Further, the program may be provided through an electric communication line such as the Internet.

(Embodiment 2)
In the configuration example described above, the air conditioner 30 is assumed as a control target, and a control plan for the air conditioner 30 is generated using a template in which a user ID is associated with a cluster ID and associated with a cluster ID. The air conditioner 30 can perform not only the cooling operation but also the heating operation, and the control target is not limited to the air conditioner 30. Therefore, it is desirable that the control plan not be uniform throughout the year, but to change according to other factors that change the environment, such as the time of day, the day of the week, the weather, the season, etc.

  In the present embodiment, as in the first embodiment, it is assumed that the control target is the air conditioner 30, and attention is paid to the outside temperature as another factor that changes the environment. The outside temperature varies depending on the time of day, weather, and season. The environmental elements controlled by the air conditioner 30 include humidity as well as temperature. Therefore, it is possible to use the humidity of the outside air when generating the control plan. However, in order to simplify the explanation, an operation that does not consider humidity will be described, and then a case where humidity is considered will be described. As the outside air temperature, data measured by a temperature sensor arranged outdoors is used, or weather data provided through an electric communication line such as the Internet is used. The temperature sensor arranged outside the room may be a temperature sensor built in the outdoor unit of the air conditioner 30. In the present embodiment, it is assumed that the sensor 20 shown in FIG. 1 includes a temperature sensor that measures the outside air temperature.

  The outside air temperature is taken into the environment control device 10 at a cycle of about 10 minutes to 1 hour, for example, and accumulated in the storage unit 15 over one to a plurality of years. The storage unit 15 stores the outside air temperature data together with the control amount of the air conditioner 30 in association with the date and time. Therefore, it is possible to associate the change in the control amount of the air conditioner 30 over time with the outside air temperature. Further, the change in the control amount of the air conditioner 30 is also associated with the date and time.

  Therefore, by using the information stored in the storage unit 15, the change in the control amount of the air conditioner 30 over time depends on the season and the user's life pattern, not the user's emotional characteristics in the specific environment. It will be associated with the emotional features. As a result, for example, it is possible to associate a change in emotional characteristics due to a user's biological rhythm with a template, and it is possible to set a highly accurate control plan. The template is updated at an appropriate timing. The template is updated by the user or automatically at a predetermined timing.

  In order to create a template from the information stored in the storage unit 15, the analysis unit 16 extracts a control pattern (see FIG. 3) of the air conditioner 30 over time for each temperature difference between the outside air temperature and the room temperature. . The temperature difference is a predetermined value such as 3 ° C. and 5 ° C. (may include 1 ° C.). If the temperature difference is not these values, round off to the values. Rounded. The analysis unit 16 obtains the frequency of the control pattern for each temperature difference, and extracts the control pattern with the maximum frequency.

  When a plurality of users share the air conditioner 30 like the air conditioner 30 shared by the family in the house, the control pattern may differ depending on the sensibility characteristics of each user. Even in this case, as described above, by extracting the control pattern having the maximum frequency, a control pattern that is considered to be agreed between a plurality of users is extracted.

  Similarly to the temperature change pattern, the target environment and control time related to temperature are also tabulated for each temperature difference, and a value with a high occurrence frequency is selected. Here, the target environment and control time related to temperature are not the frequency of occurrence of individual values, but the frequency of occurrence is determined for each section of a certain width.

  In addition, when generating the control plan of the air conditioner 30, in addition to adopting information for each day, information obtained by averaging information for a plurality of days may be employed. However, a plurality of days in which information is adopted to generate a control plan may include days whose outside air temperature is significantly different from other days. Therefore, the difference between the average value of the outside temperature on multiple days and the outside temperature on each day is calculated, and if this difference is equal to or greater than a predetermined reference value, the corresponding information is information for generating a control plan. It is desirable to be excluded from. For example, if information on a day that is a difference of 7 ° C. or more with respect to the average value of the outside air temperature is excluded from the information for generating the control plan, the validity of the averaged information becomes high. When a plurality of types of control plans are generated for the same outside air temperature, the control plan to be adopted may be determined by using other information such as humidity together.

  According to the configuration of the present embodiment, since information accumulated over one to a plurality of years is used, a control plan can be generated for each season. In particular, when information for a plurality of years is used, it is possible to reflect the difference in the environment for each year in the control of the air conditioner 30. For example, by considering the characteristics of each year such as a year with severe heat, a cold summer year, a warm winter year, and a long rainy season, it is possible to generate a control plan that matches the relevant characteristics.

  Even if the room temperature and the outside air temperature are the same, the degree of adaptation of the living body is different and the clothes are different. For example, between the warm day in January and the severely chilled day in March, there is a difference in the control content of heating by the air conditioner 30 depending on the environment, clothes, living body adaptation, etc. for several days before the corresponding day. It can happen. By storing not only the outside air temperature but also the control content of the air conditioner 30 in the storage unit 15 in association with the date and time, when the outside air temperature alone cannot narrow down the control plan, the outside air temperature change pattern is considered. This makes it possible to generate a control plan that takes into account the characteristics of each year.

  It is also possible to automatically determine the timing of the cooling operation and the heating operation by reflecting the user's habits according to the outside air temperature by associating the control contents of the past plural years with the date and time. .

  By the way, in order to generate the control plan of the air conditioner 30 according to the life pattern, it is necessary to classify the operation time and non-operation time of the air conditioner 30 in one day for each day with different life patterns. On days with different life patterns, weekdays and holidays are different, and when there are children, it is necessary to consider summer and winter holidays. In order to classify days with different life patterns, it is desirable to pay attention to the time at which a pattern for each user is likely to appear, such as wake-up time, return time, and bedtime. Since there is an error in these times, it is desirable to use a time zone having a width of several minutes to 30 minutes.

  When using the time of such a life pattern, the control plan may be generated so as to correspond to a life pattern in which the coincidence rate of each time is a reference value (for example, 80%) or more. It should be noted that the standard values for the time zone width and the coincidence rate are set as standard values as default values, and it is desirable that the user set or adjust appropriately as necessary.

  In the present embodiment, the control plan is generated in consideration of the influence of the external environment, such as the outside air temperature. The environment controlled by the air conditioner 30 such as room temperature is affected by the outside air temperature, but is also affected by the weather such as the amount of solar radiation. Therefore, it is desirable to add weather in addition to the outside temperature as information on the external environment. That is, it is desirable to store the weather information in the storage unit 15 as well as the outside air temperature information. Weather information can be obtained through a telecommunication line such as the Internet, for example, at a cycle of about 10 minutes to 1 hour.

  When the planning unit 13 generates the control plan, it acquires the weather forecast information through the telecommunication line, and collates it with the weather information stored in the storage unit 15. It is possible to generate a control plan in consideration of the weather. That is, it is possible to generate a control plan in consideration of not only the outside air temperature but also the weather.

  As in the first embodiment, the control plan generated as described above can be evaluated by the user using the operation display 40, and the user can evaluate the control plan generated by the environment control apparatus 10. can get. In other words, if the user is dissatisfied with the control plan, this evaluation provides feedback reflecting the user's emotional characteristics. When such feedback is obtained, the control plan is corrected or changed in the same manner as in the first embodiment, and a control plan that is finally satisfied by the user is adopted. In short, the control plan automatically generated by the environmental control device 10 so as to suit the cluster to which the user belongs is adjusted as necessary to suit the emotional characteristics of each user, and finally, the individual users It is possible to generate a satisfactory control plan. Other configurations and operations are the same as those of the first embodiment.

  The above-described embodiment is an example of the present invention. For this reason, the present invention is not limited to the above-described embodiment, and various modifications can be made according to design and the like as long as the technical idea according to the present invention is not deviated from this embodiment. Of course, it can be changed.

DESCRIPTION OF SYMBOLS 10 Environmental control apparatus 11 Input part 12 Output part 13 Planning part 14 Execution part 15 Storage part 30 Air-conditioning equipment (equipment)

Claims (4)

A storage unit that stores a control pattern that is a change in a control amount with the passage of time in association with a cluster that classifies emotional features for each user, and
An input section for inputting an evaluation for the environment;
A planning unit that generates a control plan using the control pattern extracted from the storage unit using an emotional feature based on information relating to the evaluation given to the input unit;
An environment control apparatus comprising: an execution unit that controls equipment that controls the environment according to the control plan generated by the plan unit. The input unit is configured to receive an evaluation for the environment during a period in which the device is controlled according to the control plan,
The environment control device according to claim 1, wherein the planning unit is configured to correct or change the control plan when the evaluation indicates dissatisfaction. The environment control device according to claim 1, wherein the storage unit is configured to preferentially accumulate a control pattern in which the evaluation of the user is greater than or equal to a reference value.   A program that causes a computer to function as the environment control device according to any one of claims 1 to 3.

Images