JP7575356B2 - Maintenance Management System - Google Patents

Description

The present invention relates to a plant equipment maintenance and management system that maintains and manages plant equipment such as chemical plants, power plants, water and sewage plants, and waste disposal plants.

In the maintenance and management of plant equipment such as chemical plants, power plants, water and sewage plants, and waste disposal facilities, for example, if an operator hears an abnormal sound that is not normally present, the plant equipment may be shut down, and maintenance and management of the plant equipment may be dependent on the operator's five senses and experience.

When a plant facility operates continuously 24 hours a day, operators may need to be present at the facility at all times to handle maintenance work. However, Japan's productive-age population (15-64 years old) is on the decline, and if a labor shortage occurs in the future, this type of response may become difficult.

In order to continue such maintenance work even in the event of a labor shortage, the five senses and experience of operators will be replaced with IoT (Internet of Things) technology and AI (artificial intelligence) technology. For example, cameras will be installed on plant equipment to replace the operator's vision, and image and video data captured by the cameras will be processed to detect changes in the appearance of the plant equipment. Also, microphones will be installed on plant equipment to replace the operator's hearing, and audio data collected by the microphones will be processed to detect abnormal sounds from the plant equipment.

In addition to the operator's sight and hearing, the operator's sense of smell is also important for detecting strange odors (such as burnt or foul odors) that occur in plant facilities. However, the threshold for detecting strange odors varies from person to person, and there are individual differences, so this becomes tacit knowledge for each operator, making it difficult to pass on knowledge.

Therefore, odor sensors and concentration sensors capable of measuring the concentration of odorous substances are installed in plant equipment to replace the operators' sense of smell and detect abnormal odors occurring in the plant, which are then used for maintenance work.

JP 2014-209307 A (hereinafter referred to as Patent Document 1) is an example of background technology in this technical field.

Patent Document 1 describes a plant equipment operation system having an event information forming means for extracting events that have occurred in the plant equipment, a response information forming means for extracting responses taken in response to events by experienced operators with a certain level of ability, a storage means for storing the events extracted by the event information forming means and the responses extracted by the response information forming means in association with each other, a current event information forming means for extracting events currently occurring in the plant equipment, and a response information selecting means for selecting from the storage means an event that matches the event extracted by the current event information forming means.

Patent document 1 also describes how the audio recording of "odor" uttered by an experienced operator can be acquired.

JP 2014-209307 A

The operation system described in Patent Document 1 (a maintenance management system for plant equipment) associates the vocalization of "odor" made by a skilled operator with an event that occurs in the plant equipment.

However, the "smelly" sound alone is unclear as to whether it is expressing the strength of the odor or the pleasantness of the odor, such as whether it is pleasant or unpleasant, and while an experienced operator may be able to understand it, other operators may not be able to understand it.

The present invention provides a maintenance management system for plant equipment that can determine the operating status of plant equipment even if other operators do not have the know-how by making the knowledge about odors, which has previously been implicit knowledge of each operator, explicit knowledge.

In order to solve the above-mentioned problems, the maintenance management system of the present invention includes a five- sense sensor information acquisition means for acquiring five-sense sensor information, which is an odor level obtained by converting a current value measured by an odor sensor that is installed in a plant facility and measures the concentration of odor-causing substances and displays it as a numerical value, into the concentration of odor-causing substances and the time when the five-sense sensor information was recorded , a sensory information acquisition means for acquiring sensory information, which is odor information that is a step-rating of evaluation items confirmed by an operator on-site, such as odor strength, odor comfort, odor acceptability, and odor impression, and the time when the sensory information was recorded , and a five-sense sensor information acquisition means for acquiring the sensory information acquired at the same time based on the time when the five-sense sensor information was acquired and the time when the sensory information was acquired. the five senses sensor information and the sensory information acquired by the sensory information acquisition means; a plant information acquisition means for acquiring plant information, which is the operating state of the plant equipment , and the time when the plant information was recorded ; a plant normal/abnormal discrimination means for discriminating whether the operating state of the plant equipment is normal or abnormal based on the plant information; a sensory information/plant information integration means for integrating the five senses sensor information and the sensory information integrated at the same time , the plant information acquired at the same time, and the discrimination result based on that time; and a database in which the five senses sensor information, the sensory information, the plant information, and the discrimination result are stored based on that time .

According to the present invention, by making the knowledge about odors that was previously implicit to each operator explicit, it is possible to provide a maintenance management system for plant equipment that can determine the operating status of the plant equipment even if other operators do not have the know-how.

Note that issues, configurations and effects other than those mentioned above will become clearer in the explanation of the examples below.

FIG. 1 is an explanatory diagram illustrating a maintenance management system in a first embodiment. FIG. 2 is an explanatory diagram illustrating the odor evaluation items in Example 1. FIG. 2 is an explanatory diagram illustrating a database that integrates sensory information and plant information in the first embodiment. FIG. 2 is an explanatory diagram for explaining the maintenance management using odor in the first embodiment. FIG. 11 is an explanatory diagram for explaining a maintenance management system in a second embodiment. FIG. 13 is an explanatory diagram for explaining a maintenance management system in a third embodiment. FIG. 11 is an explanatory diagram illustrating the relationship between odor intensity and odor level for each operator in Example 3.

Below, an embodiment of the present invention will be described with reference to the drawings. Note that the same reference numerals are used in each drawing for substantially the same or similar configurations, and where explanations are repeated, they may be omitted.

First, we will explain the maintenance management system in Example 1.

Figure 1 is an explanatory diagram explaining the maintenance management system in the first embodiment.

The plant equipment maintenance management system described in the first embodiment has the following configuration.
(1) A five-sensory sensor information acquiring means 100 that acquires five-sensory sensor information measured by a five-sensory sensor installed in a plant facility.
(2) A sensory information acquisition means 200 that acquires sensory information (operator's sensory information) confirmed by the operator at the site.
(3) A sensory information/sensor information integration means 300 that integrates (combines) the five-sensor information acquired by the five-sensor information acquisition means 100 and the sensory information acquired by the sensory information acquisition means 200.
(4) A plant information acquiring means 400 for acquiring plant information which is the operating state of the plant equipment.
(5) Plant normal/abnormal discriminating means 500 for discriminating whether the operating state of the plant equipment is normal or abnormal based on the plant information.
(6) A sensory information/plant information integration means 600 that integrates (combines) the integrated five-sensory sensor information and sensory information with plant information having discrimination results (integrating the five-sensory sensor information, sensory information, discrimination results, and plant information).

The communication means connecting each means is not particularly limited and may be a wireless LAN (Local Area Network) or a wired LAN.

The maintenance management system is stored in a computer having a computing means such as a CPU (Central Processing Unit) (not shown), a storage means such as a memory or a hard disk, and an input/output means such as an interface, and is preferably stored as software.

In the first embodiment, the maintenance management system is stored as software in a computer having a calculation means, a storage means, and an input/output means, but the maintenance management system does not necessarily have to be installed in the plant equipment and may be installed in a location away from the plant equipment.

In addition, the sensory information/sensor information integration means 300 and the sensory information/plant information integration means 600 may be integrated to integrate the five senses sensor information, sensory information, discrimination results, and plant information together.

The five-sense sensor information acquisition means 100 is a five-sense sensor installed in plant facilities such as chemical plants, power plants, water and sewage plants, and waste disposal plants, and acquires five-sense sensor information from the plant facilities.

Here, the five senses sensors are sensors that substitute for human senses, such as cameras that acquire image data and video data (including infrared cameras that visualize heat), microphones that acquire audio data, odor sensors that measure odors, and concentration sensors that can acquire the concentration of odorous substances.

The five senses sensor may be a fixed sensor or a mobile (handheld) sensor. When a mobile sensor is used, even if there are multiple locations from which the five senses sensor information is acquired, a single mobile sensor can be used. When a mobile sensor is used, a GPS (Global Positioning System) function is also added to the mobile sensor, and the location from which the five senses sensor information was acquired is also acquired along with the five senses sensor information.

Here, the five senses sensor information refers to image data or video data acquired by a camera, audio data acquired by a microphone, odor levels acquired by an odor sensor, and odorant concentrations acquired by a concentration sensor capable of acquiring the concentration of odorant substances.

It should be noted that image data, video data, and audio data may be pre-processed. Specifically, image data and video data may be subjected to image processing or video processing to extract brightness information, color information, and pattern information of specific locations, which may then be used as sensor information for the five senses. Audio data may be subjected to FFT (fast Fourier transform) processing to extract power spectrum values for each frequency, and an overall value for all frequencies may then be extracted, which may then be used as sensor information for the five senses. In other words, the information necessary for plant equipment maintenance and management work may be obtained from various pieces of information.

In Example 1, we will explain the odor from a sewage treatment plant in a water supply and sewerage plant as an example.

A sewage treatment plant is a plant facility that purifies wastewater from a sewer system and discharges the treated water into rivers, lakes, or the sea, and is made up of multiple plant facilities. For example, a sewage treatment plant may have multiple plant facilities, such as a grit basin, a primary sedimentation basin, a reaction tank, a final sedimentation basin, a sludge thickening tank, and a sludge dewatering facility.

Wastewater is treated in these multiple plant facilities, and during the wastewater treatment process, odor-causing substances are dispersed from the wastewater and treated water and released into the atmosphere.

The main substances that cause odors (odor components) are hydrogen sulfide, methyl mercaptan, methyl sulfide, methyl disulfide, chlorine, and ammonia. At sewage treatment plants, these odor-causing substances (bad odors) are collectively treated as odors.

In addition, in the case of water purification plants, when treated water is disinfected as a measure against viruses and bacteria, the chlorine used for disinfection (a substance that causes odors) can also cause odors such as a chlorine smell or a chlorine odor, and the treated water can also have a moldy smell (geosmin, 2-MIB), earthy smell, or algae smell.

In Example 1, an odor sensor is used as the five senses sensor, and odor sensor information, which is information on substances that cause odors, is obtained as the five senses sensor information.

The odor sensor measures the concentration, index, and intensity of odor-causing substances and displays them as numerical values. The odor sensor information is odor sensor measurement information, specifically, odor concentration, odor index, odor intensity, and odor level indicated by the odor sensor. In Example 1, odor sensor information (e.g., odor level) is used as the five senses sensor information. Note that the odor level is a value indicated by the odor sensor, and is the current value acquired by the odor sensor converted into the concentration of odor-causing substances. In Example 1, in particular, the odor level of a reaction tank (the location where the five senses sensor information is acquired) in a sewage treatment plant is acquired.

Various odor sensors are in practical use or under research and development, including semiconductor types that attach odor-causing substances to the surface of a semiconductor and use an oxidation-reduction reaction; chemical types that use an adsorption film to which odor-causing substances are adsorbed; photoacoustic spectroscopy types that irradiate odor-causing substances with light of a specific wavelength and measure vibrations; and biological types that use olfactory cells from humans, animals, and insects.

The odor sensor used in Example 1 is not particularly limited as long as it can measure the main odor-causing substances in sewage treatment plants, namely hydrogen sulfide, methyl mercaptan, methyl sulfide, methyl disulfide, chlorine, and ammonia, as odors.

In addition, for plant facilities other than sewage treatment plants, an odor sensor that can measure odors generated in each plant facility can be used.

The sensory information acquisition means 200 acquires sensory information observed by the operator on-site.

Here, sensory information refers to information obtained from the operator's five senses, including sight, hearing, and smell. Here, the site refers to the location where the plant equipment is installed, for example, the location where a reaction tank is installed in a sewage treatment plant.

For example, in a reaction tank (where sensory information is obtained) at a sewage treatment plant, sensory information includes, for vision, the height of the treated water liquid level, the color of the sludge, and the appearance of bubbles due to aeration; for hearing, the operating sound of the mixing device that mixes the treated water; and for smell, odor information includes odor intensity, odor comfort (pleasant/unpleasant), odor acceptability, and odor impression (type).

In Example 1, the sensory information acquisition means 200 uses a recording device such as a tablet (mobile information terminal) that an operator can operate by touching the screen to record sensory information (odor information) observed on-site, for example, regarding odors in reaction tanks at sewage treatment plants, and acquires this recorded sensory information (odor information).

When sensory information is recorded using a recording device such as a portable information terminal, the time (date and time) when the operator recorded the sensory information is recorded together with the sensory information, and the recorded sensory information and time are transmitted to the sensory information/sensor information integration means 300.

For example, when sensory information is recorded using a handwritten record sheet or the like, the sensory information recorded in the handwritten record sheet or the like is input from an interface such as a keyboard together with the time when the sensory information was recorded, and the input sensory information and time are transmitted to the sensory information/sensor information integration means 300.

The sensory information/sensor information integration means 300 integrates the five-sensory sensor information acquired by the five-sensory sensor information acquisition means 100 and the sensory information acquired by the sensory information acquisition means 200. Then, the five-sensory sensor information and sensory information acquired at the same time are integrated based on that time (as a key). The integrated information is then stored in a database that exists in a memory means and is associated with the sensory information/sensor information integration means 300 and the sensory information/plant information integration means 600. The integrated information may be output (visualized) using an interface such as a monitor.

The plant information means 400 acquires plant information, which is the operating state of the plant equipment at the same time. In other words, the plant information means 400 acquires plant information, such as plant data and plant operation signals, from a large number of sensors installed in the plant equipment.

Here, the plant data includes the water temperature, pH, redox potential, total organic carbon (TOC), dissolved oxygen concentration, sludge concentration, flow rate, aeration air volume, etc. in the reaction tank, and the plant operation signals include the set values or control values of control signals such as the amount of chemicals added to the reaction tank and the valve opening degree.

The plant information also includes alarm logs that are issued when an abnormality occurs in the plant equipment (for example, a minor failure of the sludge pump, a major failure of the aeration blower, etc.).

In Example 1, plant information is acquired specifically from a reaction tank in a sewage treatment plant (where plant information is acquired).

The plant normal/abnormal discrimination means 500 discriminates whether the operating state of the plant equipment is normal or abnormal based on the plant information. In other words, the plant normal/abnormal discrimination means 500, for example, discriminates whether the acquired plant data (plant information) exceeds a threshold value for the plant data (plant information) that is set in advance, and discriminates whether the operating state of the plant equipment is normal or abnormal.

Then, the judgment result of the operating status of the plant equipment (normal or abnormal) is stored in the database together with the acquired plant data (plant information). In particular, in the case of an abnormality and an alarm log has been issued, the content of the alarm log (e.g., serious aeration blower failure) and the content of the abnormality (e.g., aeration blower abnormality) are stored in the database. In other words, the plant information and the judgment result are stored in the database.

The sensory information/plant information integration means 600 integrates the integrated five-sensory sensor information and sensory information for the same time with plant information having the discrimination result for the same time. In other words, the sensory information/plant information integration means 600 integrates the five-sensory sensor information and sensory information for the same time integrated by the sensory information/sensor information integration means 300, and the plant information and discrimination result in the plant normal/abnormal discrimination means 500 for the same time based on that time (as a key). In this way, the sensory information/plant information integration means 600 integrates the five-sensory sensor information, sensory information, discrimination result, and plant information for the same time. Then, it stores them in a database.

In other words, the database stores the five-sense sensor information and sensory information integrated by the sensory information/sensor information integration means 300, as well as the plant information and discrimination results integrated by the sensory information/plant information integration means 600, at the same time.

Next, we will explain the odor evaluation items in Example 1.

Figure 2 is an explanatory diagram explaining the odor evaluation items in Example 1.

In Example 1, when an operator checks odor information on-site, the operator uses a recording device such as a portable information terminal to evaluate the odor information, including odor strength, odor comfort (pleasant/unpleasant), odor acceptability, and odor impression (type).

In Example 1, these four evaluation items are evaluated on a six-level scale (0 to 5). For example, the odor intensity is evaluated on six levels: "odorless," "weak," "easily detectable," "slightly strong," "strong," and "intense." The pleasantness of the odor is evaluated on six levels: "comfortable," "slightly comfortable," "normal," "slightly unpleasant," "unpleasant," and "very unpleasant." The acceptability of the odor is evaluated on six levels: "acceptable," "acceptable with some resistance," "acceptable if tolerated," "unacceptable," "very unacceptable," and "-." The impression (type) of the odor is evaluated on six levels: "rotten egg odor (hydrogen sulfide)," "rotten onion odor (methyl mercaptan)," "pungent odor (ammonia)," "disinfectant odor (chlorine)," "burnt odor," and "other."

As for the evaluation items (four items) that are used to evaluate odor information (sensory information) and the evaluation stages (six levels) that are used to evaluate the evaluation items, the evaluation items and evaluation stages can be set based on the odor information generated by each plant facility.

In this way, in Example 1, the sensory information acquisition means 200 acquires odor information evaluated by the operator on-site as evaluation items, which are items for evaluating the odor information, and evaluation stages, which are levels for evaluating the evaluation items.

In the first embodiment, the relationship between these evaluation items and evaluation stages is displayed on a recording device such as a portable information terminal, and the operator can record the odor information observed on-site. This allows the odor information to be quantified in terms of the evaluation stage for the evaluation item. The sensory information acquisition means 200 can then acquire the quantified odor information.

In Example 1, ambiguity in the sense of smell can be avoided by defining odor information in terms of evaluation items (odor intensity, odor pleasantness, odor acceptability, odor impression (type)), i.e., by defining the sense of smell.

In this way, in Example 1, by evaluating odor information using evaluation items and evaluation stages, the value of the information is increased, and the tacit knowledge of each operator regarding odors can be made explicit. In addition, the tacit knowledge of experienced operators regarding odors can be made explicit, and the operating state of the plant equipment can be determined even if other operators do not have the know-how.

Next, we will explain the database that integrates sensory information and plant information in Example 1.

Figure 3 is an explanatory diagram explaining the database that integrates sensory information and plant information in Example 1.

The database stores sensor information, sensory information, discrimination results, and plant information based on time (date and time), and stores this information as time-series trend data.

In Example 1, the five senses sensor information includes odor level, the sensory information includes odor intensity, odor comfort, odor acceptability, and odor impression (type), and the plant information includes plant information (flow rate, dissolved oxygen concentration, water temperature) and alarm logs for the location (reaction tank) where the plant information was acquired.

In addition, this information may also include other sensory information, sensory information, and plant information. It may also include information on things other than the reaction tank.

In addition, in Example 1, for convenience of explanation, the odor sensor was installed only in the reaction tank and this information was acquired, but odor sensors may be installed in multiple locations other than the reaction tank. In this case, when acquiring sensory information, it is necessary to simultaneously acquire the installation location of the odor sensor (e.g., the reaction tank).

In other words, it is preferable to unify the locations for acquiring the five-sense sensor information, the sensory information, and the plant information, and to acquire the five-sense sensor information, the sensory information, and the plant information at the same time.

Here, No. 1 is information on the date and time of AA month BB day cc hour dd minute, with the odor level being "200", the odor strength being "3", the odor comfort being "3", the odor acceptability being "2", the odor impression (type) being "1", the flow rate being "2000 ^m3 /h", the dissolved oxygen concentration being "3.5 mg/L", the water temperature being "25°C", the alarm being "-", and the discrimination result being normal.

On the other hand, No. 2 is information on the date and time EE month FF day gg hour hh minute, with the odor level being "15", the odor strength being "1", the odor comfort being "0", the odor acceptability being "1", the odor impression (type) being "1", the flow rate being "1800 ^m3 /h", the dissolved oxygen concentration being "0.5 mg/L", the water temperature being "25°C", and the alarm being "serious aeration blower malfunction", and the discrimination result being abnormal.

In the first embodiment, the plant normal/abnormal discrimination means 500 discriminates whether the operating state of the plant equipment is normal or abnormal, for example, based on the dissolved oxygen concentration (plant information), which is plant data. In other words, the plant normal/abnormal discrimination means 500 discriminates whether the acquired dissolved oxygen concentration is below a predetermined dissolved oxygen concentration (predetermined threshold) that is set in advance, and discriminates whether the operating state of the plant equipment is normal or abnormal.

The numbers for odor intensity, odor comfort, odor acceptability, and odor impression (type) are the evaluation levels given by the operators.

As such, even if there is only a slight odor, this is not necessarily normal. Experienced operators understand, through their implicit knowledge of odors, that even if there is only a slight odor, the amount of aeration is low, which reduces the dissolved oxygen concentration, and that this is caused by a malfunction of the aeration blower, meaning there is a risk of there being only a slight odor.

According to the first embodiment, the five senses sensor information, sensory information, and plant information are integrated, and when the sensory information is low, the amount of aeration is low, causing a drop in the dissolved oxygen concentration, and this can be caused by a malfunction of the aeration blower, allowing even other operators who do not have the know-how to understand that the sensory information may be low. In this way, according to the first embodiment, the implicit knowledge about odors is made explicit, allowing other operators to determine the operating status of the plant equipment.

In this way, by accumulating this information, i.e., by building such a database, odors can be used in the maintenance and management of plant facilities.

The acquired information may also be used to perform correlation analysis, which is a type of statistical analysis, to perform correlation analysis between plant information and sensory information and/or sensory information to create a model. Also, neural network technology or data clustering technology may be used to perform machine learning to create a model that detects signs of abnormalities. There are no particular limitations on how the acquired information may be used.

Next, we will explain the maintenance using odor in Example 1.

Figure 4 is an explanatory diagram explaining the maintenance using odor in Example 1.

The upper graph in Figure 4 shows the change in odor level over time, and the lower graph in Figure 4 shows the change in odor intensity over time.

In FIG. 4, when the odor level and odor intensity in a reaction tank in a sewage treatment plant fall below a preset control value, an alarm is output to indicate that a malfunction has occurred in the aeration blower. In other words, when the odor level falls below the control value and the operator's evaluation of the odor intensity (evaluation item) falls from evaluation level "3" (control value) to evaluation level "2," an alarm is output.

This allows operators to check the reaction tank based on the odor level and odor intensity. Also, if there is a discrepancy between the odor level movement and the odor intensity movement, the odor sensor can be checked. In this way, odors can be used for plant equipment maintenance work.

In particular, there are individual differences in the sense of smell, which is often dependent on the five senses of operators, and there are individual differences in how the sense of smell is perceived, making it tacit knowledge for each operator and difficult to pass on knowledge. Thus, according to Example 1, even if there are individual differences between the sense of smell of one operator and the sense of smell of another operator, the tacit knowledge regarding odors is made explicit based on the odor level and odor intensity, making it possible to determine the operating state of the plant equipment.

In addition, in Example 1, by integrating odor sensor information (e.g., odor level) from the odor sensor with sensory information from the operator (e.g., odor strength), the sensory information (judgment that varies from person to person and is ambiguous) is quantified, and further by integrating plant information with the quantified sensory information, the sensory information, which is the tacit knowledge of each operator, can be used to determine the operating status of the plant equipment (determining whether the plant equipment is normal/abnormal).

In addition, because it is difficult to constantly monitor odors, odor-generating plant equipment has been operated on the safe side with regard to odors. However, in Example 1, by acquiring odor sensor information and sensory information, the plant equipment can be operated optimally, and energy-saving operation of the plant equipment can be achieved.

Next, we will explain the maintenance management system in Example 2.

Figure 5 is an explanatory diagram explaining the maintenance management system in Example 2.

Compared to the maintenance management system described in the first embodiment, the maintenance management system described in the second embodiment has an environmental information acquisition means 700 for acquiring environmental information that influences the five-sensory sensor information, and an environmental impact assessment means 800 for evaluating the impact of the environmental information on the five-sensory sensor information acquired by the five-sensory sensor. In other words, the maintenance management system described in the second embodiment has a function of judging the validity of the five-sensory sensor information acquired by the five-sensory sensor.

The environmental information acquisition means 700 acquires sensor information that measures environmental information such as wind speed and wind direction that affect odor sensor information. Environmental information on wind speed and wind direction is a factor that affects the diffusion of odor-causing substances, and an anemometer with built-in wind direction observation means is used. An anemometer and a wind vane may also be used together.

Other environmental information besides wind speed and direction includes humidity and temperature. For example, when humidity is high, odor-causing substances tend to remain in the air. For this reason, it is preferable to install a hygrometer and thermometer to measure environmental information such as humidity and temperature.

In addition, if deodorizing fans or other equipment are installed in the plant facilities, the operation status of the fans may also have an impact.

In this way, it is preferable to define all environmental information that affects odors in the space where the odor sensor acquires odor sensor information as environmental information, and to acquire at least one piece of environmental information.

The environmental impact assessment means 800 uses the environmental information acquired by the environmental information acquisition means 700 to determine the validity of the five-sense sensor information acquired by the five-sense sensor, that is, evaluates the impact of the environmental information on the five-sense sensor information acquired by the five-sense sensor.

For example, a threshold is set for wind speed, and when it is determined that the wind speed is equal to or greater than the threshold, the environmental influence is large and the five senses sensor information may be inaccurate, an environmental impact label is added, and the environmental impact label is also stored in a database associated with the sensory information/sensor information integration means 300. The same applies to other environmental information.

In this way, by determining the validity of the sensor information acquired by the sensor, inaccurate sensor information can be eliminated when performing statistical analysis, improving the accuracy of the model.

According to the second embodiment, by having an environmental information acquisition means 700 and an environmental impact assessment means 800, the value of the five senses sensor information is increased, and the operating state of the plant equipment can be determined with even greater accuracy.

Next, we will explain the maintenance management system in Example 3.

Figure 6 is an explanatory diagram explaining the maintenance management system in Example 3.

The maintenance management system described in Example 3, compared to the maintenance management system described in Example 1, has a sensory test condition acquisition means 900 that acquires the conditions of the operator's sensory test.

The sensory test condition acquisition means 900 acquires the identification number (e.g., name or employee number) of the operator who confirmed the sensory information (e.g., odor information) at the site, that is, the conditions of the operator's sensory test. The identification number is also stored in a database associated with the sensory information/sensor information integration means 300. This makes it possible to visualize the differences in the sense of smell of each operator.

Next, we explain the relationship between odor intensity and odor level for each operator in Example 3.

Figure 7 is an explanatory diagram illustrating the relationship between odor intensity and odor level for each operator in Example 3.

For example, if an odor level equivalent to the control value is occurring, when Operator A, Operator B, and Operator C each check the odor on-site, there will be individual differences in how they perceive the strength of the odor, and Operator A may give an evaluation level of "2," Operator B a level of "3," and Operator C a level of "4."

For example, if the odor level has not been obtained and the operator's assessment of the odor intensity (assessment item) has fallen from assessment level "3" (control value) to assessment level "2," and a specified action is to be taken, operator B and operator C may not be able to take the specified action.

For example, even if the odor level has not been obtained and operator B is an experienced operator who judges the evaluation level to be "3" and is able to take the prescribed action, operator A may take the prescribed action when the odor level is higher than the control value, and operator C may take the prescribed action when the odor level is lower, which may prevent optimal maintenance work from being performed.

Therefore, by providing each operator with quantitative feedback based on the odor level, it is possible to improve the maintenance management capabilities of the operators and standardize the maintenance management capabilities of each operator.

For example, operator A is instructed to take a specific action when the evaluation stage is "2" based on his/her personal sense, operator B is instructed to take a specific action when the evaluation stage is "3" based on his/her personal sense, and operator C is instructed to take a specific action when the evaluation stage is "4" based on his/her personal sense.

In other words, in the third embodiment, the sensory test condition acquisition means 900 also determines the relationship between each operator's sensory information (e.g., odor strength) and the five senses sensor information (e.g., odor level) and stores the relationship in the database.

In this way, the manager of each operator can give appropriate instructions based on each operator's personal intuition, enabling optimal maintenance work to be carried out.

According to Example 3, by having the sensory test condition acquisition means 900, the value of the sensory information is increased, and the operating state of the plant equipment can be determined with even greater accuracy.

By storing information about the operator's physical condition (such as colds) as part of the sensory test conditions, it is possible to visualize the differences in each operator's sense of smell according to their individual circumstances.

The maintenance management system described in Example 3 preferably includes an environmental information acquisition means 700 and an environmental impact assessment means 800. This increases the value of the five senses sensor information and the value of the sensory information, and allows the operating status of the plant equipment to be determined with even greater accuracy.

The present invention is not limited to the above-mentioned embodiment, but includes various modifications. For example, the above-mentioned embodiment is specifically described in order to explain the present invention in an easy-to-understand manner, and is not necessarily limited to having all of the configurations described.

It is also possible to replace part of the configuration of one embodiment with part of the configuration of another embodiment. It is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to delete part of the configuration of each embodiment, add part of another configuration, and replace it with part of another configuration.

100: Five senses sensor information acquisition means 200: Sensory information acquisition means 300: Sensory information/sensor information integration means 400: Plant information acquisition means 500: Plant normal/abnormal discrimination means 600: Sensory information/plant information integration means 700: Environmental information acquisition means 800: Environmental impact assessment means 900: Sensory test condition acquisition means

Claims (6)

a five-sense sensor information acquiring means for acquiring five-sense sensor information , which is an odor level obtained by converting a current value measured by an odor sensor that is installed in the plant facility and measures and displays the concentration of odor-causing substances as a numerical value, and the time when the five-sense sensor information was recorded ;
a sensory information acquiring means for acquiring sensory information, which is odor information obtained by rating the evaluation items of odor intensity, odor comfort, odor acceptability, and odor impression confirmed by an operator on-site, and the time when the sensory information was recorded;
a sensory information/sensor information integration means for integrating the five-sensor information acquired by the five-sensor information acquisition means and the sensory information acquired by the sensory information acquisition means, both of which are acquired at the same time, based on the time at which the five-sensor information was acquired and the time at which the sensory information was acquired;
a plant information acquisition means for acquiring plant information, which is an operating state of the plant equipment , and a time when the plant information was recorded ;
a plant normal/abnormality determination means for determining whether an operating state of the plant equipment is normal or abnormal based on the plant information;
a sensory information/plant information integration means for integrating the five-sense sensor information and the sensory information integrated at the same time , the plant information acquired at the same time, and a discrimination result thereof , based on the time ;
a database in which the five senses sensor information, the sensory information, the plant information, and the discrimination result are stored based on the time;
A maintenance management system comprising: The maintenance management system according to claim 1 ,
The maintenance management system is characterized in that the plant equipment is a reaction tank in a sewage treatment plant. The maintenance management system according to claim 1,
A maintenance management system comprising: an environmental information acquisition means for acquiring environmental information that influences the five-sense sensor information; and an environmental impact assessment means for evaluating the impact of the environmental information on the five-sense sensor information acquired by the odor sensor. The maintenance management system according to claim 1,
A maintenance management system comprising: a sensory test condition acquisition means for acquiring conditions for a sensory test performed by an operator. The maintenance management system according to claim 4 ,
A maintenance management system, characterized in that the sensory test condition acquisition means determines a relationship between the sensory information of each operator and the five senses sensor information. The maintenance management system according to claim 1,
A maintenance management system characterized in that the locations for acquiring the five-sense sensor information, the location for acquiring the sensory information, and the location for acquiring the plant information are unified.

Images