Programs, methods, information processing devices, systems

Abstract

In a sensor device having multiple sensors that measure different items, the measurement items are measured accurately. [Solution] A program to be executed by a computer equipped with a processor and memory. The program causes the processor to perform the following steps: acquire measurement data from multiple sensors, each measuring a different item; estimate whether or not there is an abnormality in any of the multiple sensors based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items; and correct the measurement data measured by the sensor that is estimated to be abnormal based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items.

Description

Technical Field

[0001] The present disclosure relates to a program, a method, an information processing apparatus, and a system.

Background Art

[0002] A sensor device that can easily use functions corresponding to a plurality of measurement items with a single sensor device has been proposed (see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the technology described in Patent Document 1, for example, in a pH meter, when the asymmetric potential during calibration reaches a predetermined value, it is determined as an abnormality (deterioration). However, in such a sensor device, an abnormality can be determined only during calibration, and there is a risk that the pH cannot be accurately measured.

[0005] An object of the present disclosure is to accurately measure measurement items in a sensor device having a plurality of sensors for measuring different items.

Means for Solving the Problems

[0006] This is a program to be executed by a computer equipped with a processor and memory. The program causes the processor to perform the following steps: acquire measurement data from multiple sensors, each measuring a different item; estimate whether any of the multiple sensors are abnormal based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items; and correct the measurement data measured by the sensor that is estimated to be abnormal based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items. [Effects of the Invention]

[0007] According to this disclosure, a sensor device having multiple sensors that measure different items can accurately measure the items being measured. [Brief explanation of the drawing]

[0008] [Figure 1] This is a block diagram showing an example of the overall configuration of System 1. [Figure 2] Figure 1 is a schematic diagram showing the external appearance of the sensor device 10 as viewed from the front. [Figure 3] This is a schematic diagram showing a perspective view of the sensor device 10 shown in Figure 1. [Figure 4] Figures 2 and 3 are schematic diagrams showing the components related to the sensor probe of the sensor device 10. [Figure 5] Figures 2 and 3 show the piping diagrams for the sensor device 10. [Figure 6] Figures 2 and 3 are block diagrams showing the functional configuration of the sensor device 10. [Figure 7] This is a schematic diagram showing the configuration of the pH sensor 113. [Figure 8] This figure shows an example of the functional configuration of server 20. [Figure 9] This diagram shows the data structure of Plant Table 2021. [Figure 10] This diagram shows the data structure of installation table 2022. [Figure 11]It is a diagram showing the data structure of the plant environment table 2023. [Figure 12] It is a diagram showing the data structure of the measurement table 2024. [Figure 13] It is a diagram showing the data structure of the calibration table 2025. [Figure 14] It is a diagram showing the data structure of the model table 2026. [Figure 15] It is a diagram for explaining an example of the operation in which the sensor device 10 shown in FIG. 1 acquires information regarding calibration from the server 20. [Figure 16] It is a flowchart showing an example of the operation in which the sensor device 10 shown in FIGS. 2 and 3 performs a calibration process. [Figure 17] It is a flowchart showing an example of the operation of the sensor shown in FIG. 7. [Figure 18] It is a flowchart showing an example of the operation of the sensor device 10 shown in FIG. 6. [Figure 19] It is a flowchart showing an example of the operation of the server 20 shown in FIG. 8. [Figure 20] It is a schematic diagram showing an example of the display of the terminal device 30 used by the user. [Figure 21] It is a schematic diagram showing an example of the display of the terminal device 30 used by the user. [Figure 22] It is a schematic diagram showing an example of the display of the terminal device 30 used by the user. [Figure 23] It is a flowchart showing an example of the operation of the server 20 shown in FIG. 8. [Figure 24] It is a schematic diagram showing an example of the display of the terminal device 30 used by the user. [Figure 25] It is a flowchart showing an example of the operation of the server 20 shown in FIG. 8. [Figure 26] It is a schematic diagram showing an example of the display of the terminal device 30 used by the user. [Figure 27] It is a flowchart showing an example of the operation of the server 20 shown in FIG. 8. [Figure 28] It is a block diagram showing the basic hardware configuration of the computer 90. [Modes for carrying out the invention]

[0009] The embodiments of this disclosure will be described below with reference to the drawings. In all the drawings illustrating the embodiments, common components are denoted by the same reference numerals, and repeated explanations are omitted. The following embodiments are not intended to unduly limit the content of this disclosure as described in the claims. Not all components shown in the embodiments are necessarily essential components of this disclosure. Also, each drawing is a schematic diagram and is not necessarily a strict illustration.

[0010] <Overview> In this embodiment, the sensor device has a configuration having multiple sensors, each measuring a different item. If an abnormality occurs in any of the sensors, the sensor that caused the abnormality is estimated based on the current measurement data and past measurement data. The sensor device corrects the measurement value from the sensor that is estimated to have caused the abnormality.

[0011] <1 System Configuration> Figure 1 is a block diagram showing an example of the overall configuration of System 1. System 1, as shown in Figure 1, manages measurement data measured in, for example, a water treatment facility. System 1 includes, for example, a sensor device 10, a server 20, and a terminal device 30. The sensor device 10, server 20, and terminal device 30 communicate with each other via, for example, a network 80.

[0012] The sensor device 10 shown in Figure 1 is an information processing device that is installed at various locations in a water treatment facility and measures multiple types of items related to water at the installed location. In this embodiment, the water treatment facility may include various water treatment facilities such as groundwater utilization facilities, sewage treatment facilities, and water purification facilities. The sensor device 10 detects at least one of the following components: water supplied to the treatment equipment constituting the water treatment facility, water being treated by the treatment equipment, and water discharged from the treatment equipment.

[0013] The components of a water treatment facility include, for example, a raw water tank, prefilter, membrane filter, treated water tank, or receiving tank, if the water treatment facility is a groundwater utilization facility. If the water treatment facility is a sewage treatment facility, the components include, for example, a grit chamber, primary sedimentation chamber, reaction tank, final sedimentation chamber, or disinfection equipment. If the water treatment facility is a water purification facility, the components include, for example, an intake well, flocculation tank, sedimentation tank, filtration tank, or water purification tank.

[0014] In Figure 1, System 1 describes sensor devices 10 installed in three water treatment facilities, but the number of water treatment facilities where the sensor devices 10 are installed is not limited to three. The number of water treatment facilities where the sensor devices 10 are installed may be less than three or more than three.

[0015] Furthermore, the sensor device 10 is not limited to being installed in a water treatment facility. The sensor device 10 may be installed in a water treatment device provided in each settlement of a predetermined size, or it may be installed in an individual water treatment device.

[0016] Furthermore, the device for measuring water-related parameters is not limited to the sensor device 10. For example, System 1 may include the following devices: • A device that detects the operating status of the processor. • A device for measuring water usage • Devices for measuring pollutant emissions • A device for measuring the amount of water reused. • A device that detects power consumption in the processor.

[0017] Server 20 is an information processing device that manages data related to water treatment facilities and measurement data measured at water treatment facilities, and evaluates water treatment at a designated water treatment facility.

[0018] In this embodiment, a collection of multiple devices may be treated as a single server. The method of allocating the multiple functions required to implement the server 20 according to this embodiment to one or more hardware can be appropriately determined in view of the processing capacity of each hardware and / or the specifications required for the server 20.

[0019] The terminal device 30 is, for example, an information processing device operated by a user utilizing a service provided by the server 20. The terminal device 30 can be implemented as a stationary PC (Personal Computer) or a laptop PC, etc. The terminal device 30 may also be implemented as a mobile device such as a smartphone or tablet, for example.

[0020] Each information processing device consists of a computer equipped with an arithmetic unit and a memory device. The basic hardware configuration of the computer and the basic functional configuration of the computer realized by said hardware configuration will be described later. For each of the sensor device 10, server 20, and terminal device 30, explanations that overlap with the basic hardware configuration and basic functional configuration of the computer described later will be omitted.

[0021] <1.1 Sensor Device Configuration> Figure 2 is a schematic diagram showing the sensor device 10 shown in Figure 1 as viewed from the front. Figure 3 is a schematic diagram showing a perspective view of the sensor device 10 shown in Figure 1. Figure 4 is a schematic diagram showing the components related to the sensor probe of the sensor device 10 shown in Figures 2 and 3. Figure 5 is a piping diagram of the sensor device 10 shown in Figures 2 and 3. Figure 6 is a block diagram showing the functional configuration of the sensor device 10 shown in Figures 2 and 3.

[0022] The sensor device 10 has components for measuring water quality housed in a housing 11. The housing 11 is fitted with a lid 12 that can be opened and closed.

[0023] The housing 11 has a roughly rectangular parallelepiped shape with one open side. One side of the housing 11 (the first side) has a pipe connection hole 11a for draining sample water, a pipe connection hole 11b for supplying sample water, pipe connection holes 11c (three in the example in Figure 3) for power input and signal output, and an air intake port 11d. The side of the housing 11 opposite to the first side (the second side) has an exhaust port 11e (not shown).

[0024] A valve 1110 is installed in the pipe connection hole 11a. A valve 119 is installed in the pipe connection hole 11b. An air filter 1118 is installed in the air intake port 11d.

[0025] The housing 11 contains various sensors 111-118, a flow cell 1111, a control box 1112, a terminal block 1113 for power input, a terminal block 1114 for external output, an air vent valve 1115, a drain valve 1116, and a flow rate control valve 1117.

[0026] The various sensors 111 to 118 include, for example, a sensor 111 for measuring electrical conductivity (hereinafter referred to as an EC sensor) equipped with an electrical conductivity cell. The various sensors 111 to 118 include, for example, a sensor 112 for measuring residual chlorine concentration (hereinafter referred to as an FCL sensor). The various sensors 111 to 118 include, for example, a sensor 113 for measuring pH (hereinafter referred to as a pH sensor) equipped with electrodes (reference electrode and reference electrode) for measuring pH. The various sensors 111 to 118 include, for example, a sensor 114 for measuring oxidation-reduction potential (hereinafter referred to as an ORP sensor) equipped with electrodes (reference electrode and reference electrode) for measuring oxidation-reduction potential. The various sensors 111 to 118 include, for example, a sensor 115 for measuring nitrate ions (hereinafter referred to as a NO3 sensor). The various sensors 111 to 118 include, for example, a sensor (hereinafter referred to as a FLOW sensor) 116 for measuring the flow rate of water flowing into the sensor device 10. The various sensors 111 to 118 include, for example, a sensor (hereinafter referred to as a TUR sensor) 117 for measuring the turbidity of water flowing into the sensor device 10. The various sensors 111 to 118 include, for example, a sensor (hereinafter referred to as a TEMP sensor) 118 for measuring the temperature of water flowing into the sensor device 10.

[0027] The sensors used in the sensor device 10 are not limited to those listed above. For example, a sensor that measures other parameters may be installed in place of any of the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, and NO3 sensor 115. Alternatively, any of the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, and NO3 sensor 115 may be omitted. Furthermore, in addition to the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, and NO3 sensor 115, a sensor that measures other parameters may be installed. In this case, for example, the number of flow cells 1111 may be increased in proportion to the increase in the number of sensors. The increase in the number of flow cells 1111 may be addressed, for example, by increasing the capacity of the housing 11.

[0028] For example, sensors 111 to 115 may sense at least one of the following. In addition to the above sensors, there may be sensors that sense at least one of the following. (1) Alkalinity, ion concentration, hardness (2) Chromaticity, viscosity, dissolved oxygen (3) Odors, ammonia nitrogen, nitrate nitrogen, nitrite nitrogen, total nitrogen, total phosphorus, total organic carbon, total inorganic carbon, total trihalomethanes (4) Microbial sensor detection results, chemical oxygen demand, biological oxygen demand (5) Cyanide, mercury, oil, surfactants (6) Detection results from optical sensors and TDS (Total Dissolved Solids) sensors (7) Mass spectrometry results, fine particles, zeta potential, surface potential

[0029] Sensors 111-115 and 118 are detachably housed in the flow cell 1111. Referring to Figure 4, the shape of, for example, the NO3 sensor 115 will be described. The shapes of sensors 111-114 and 118 are similar to those of the NO3 sensor 115.

[0030] The NO3 sensor 115 has a gripping portion 115a, a flange portion 115b, and a measuring portion 115c. The gripping portion 115a is the region that protrudes from the flow cell 1111 when the NO3 sensor 115 is mounted on the flow cell 1111. The gripping portion 115a is formed in a shape that is easy for a user to grip, for example. A cord for power supply and data transmission is connected to the top of the gripping portion 115a.

[0031] The flange portion 115b is a region that acts as a stopper when the NO3 sensor 115 is inserted into the flow cell 1111. A groove is formed in a part of the flange portion 115b on the measuring portion 115c side to function as a screw.

[0032] The measuring section 115c is an area where components for measuring water quality are housed. The measuring section 115c is inserted into the measuring port 11112 formed in the flow cell 1111. The outer diameter of the measuring section 115c is smaller than the inner diameter of the measuring port 11112.

[0033] The flow cell 1111 is a component that brings the sample water into contact with the measuring sections of sensors 111-115 and 118, allowing the sensors 111-115 and 118 to accurately measure the water quality. In this embodiment, the flow cell 1111-1 is formed to accommodate sensors 111-113. The flow cell 1111-2 is formed to accommodate sensors 114, 115 and 118. In this embodiment, the flow cells 1111-1 and 1111-2 are described in a configuration that can accommodate three sensors each. However, the flow cell may be configured to accommodate three or more sensors, or fewer than three sensors.

[0034] Here, referring to Figure 4, the shape of flow cell 1111-2 will be described, for example. Note that the shape of flow cell 1111-1 is the same as that of flow cell 1111-2. Flow cell 1111-2 is formed so that the gripping portions of sensors 114, 115, and 118 face towards the opening of the housing 11. Flow cell 1111-2 is formed so that the gripping portions of sensors 114, 115, and 118 face upward. This makes it easier for the user of the sensor device 10 to house and remove sensors 114, 115, and 118. It also prevents water from leaking from flow cell 1111-2 when sensors 114, 115, and 118 are removed.

[0035] The flow cell 1111-2 has measurement ports 11111, 11112, and 11113, a sample water supply channel 11114, and an air vent channel 11115. Here, referring to Figure 4, we will describe the shape of measurement port 11112, for example. Note that the shapes of measurement ports 11111 and 11113 are the same as the shape of measurement port 11112.

[0036] The measuring port 11112 has a first cylindrical portion 11112a and a second cylindrical portion 11112b. The first cylindrical portion 11112a is formed to open at a predetermined angle toward the front of the housing 11. The inner diameter of the first cylindrical portion 11112a is formed to be approximately the same as the outer diameter of the flange portion 115b of the NO3 sensor 115. A groove is formed along the circumferential direction on the inner wall of the first cylindrical portion 11112a. This groove engages with a groove formed in the flange portion 115b, making it possible to fix the NO3 sensor 115 to the measuring port 11112.

[0037] The second cylindrical portion 11112b is formed from the bottom of the first cylindrical portion 11112a in the same direction as the first cylindrical portion 11112a. The inner diameter of the second cylindrical portion 11112b is smaller than the inner diameter of the first cylindrical portion 11112a. The inner diameter of the second cylindrical portion 11112b is slightly larger than the outer diameter of the measuring portion 115c of the NO3 sensor 115.

[0038] The second cylindrical section 11112b intersects with the water supply channel 11114 near its bottom. The second cylindrical section 11112b intersects with the air vent channel 11115 at a position closer to the front in the opening direction than the water supply channel 11114. The water supply channel 11114 intersects with the second cylindrical section 11112b on the back side of the flow cell 1111-2, closer than the air vent channel 11115. Also, the water supply channel 11114 intersects with the second cylindrical section 11112b at a higher position than the air vent channel 11115.

[0039] The second cylindrical portion 11113b of the measuring port 11113 has a hole 11113c formed at its bottom. A fitting 1122 is installed in the hole 11113c. The fitting 1122 is connected, for example, to a drain valve 1116 by a hose.

[0040] The water supply channel 11114 is formed to penetrate the flow cell 1111-2 vertically. A fitting 1120 is installed at the upper end of the water supply channel 11114. The fitting 1120 is connected by a hose to a fitting 1121 installed at the lower end of the water supply channel 11114 of the flow cell 1111-1, for example. The fitting 1121 is installed at the lower end of the water supply channel 11114. The fitting 1121 is connected by a hose to a pipe connection hole 11a and a drain valve 1116, for example.

[0041] The air vent passage 11115 is formed with the measurement port 11113 at its lower end and exits the flow cell 1111-2 upwards. An air vent valve 1115 is installed at the upper end of the air vent passage 11115. The air vent valve 1115 can be opened and closed by operating a knob.

[0042] The FLOW sensor 116 is connected to the pipe connection hole 11b via a flow control valve 1117. The flow rate can be increased or decreased by turning the flow control knob on the flow control valve 1117. The FLOW sensor 116 measures the flow rate of the sample water supplied from the pipe connection hole 11b.

[0043] The TUR sensor 117 is inserted into the shell 1171. The shell 1171 is filled with sample water supplied from the FLOW sensor 116. The sample water that has passed through the shell 1171 is supplied to the fitting 1120 located above the flow cell 1111-1.

[0044] Sensors 111 to 118 are supplied with sample water via the piping system shown in Figure 5.

[0045] The control box 1112 controls the power supply and various operations of the sensor device 10. The control box 1112 includes, for example, a circuit breaker switch, connectors for connecting to various sensors, a connector for connecting to a communication interface, a fan for CPU cooling, a connection board for transmitting data to or receiving data from various sensors 111 to 118, and a board for mounting the CPU.

[0046] A touch panel 1119 is installed on the cover 12. The touch panel 1119 includes, for example, a touch-sensitive device 11191 and a display 11192. The touch-sensitive device 11191 is an example of an input device for a user operating the sensor device 10 to input instructions or information. The touch-sensitive device 11191 receives instructions when the user touches its operating surface, accepts input from the user, and outputs the input received from the user to the control box 1112. The display 141 is an example of an output device that displays data corresponding to the control of the control box 1112. The display 141 is implemented by, for example, an LCD (Liquid Crystal Display) or an organic EL (Electro-Luminescence) display.

[0047] As shown in Figure 6, the sensor device 10 comprises a communication unit 120, a touch panel 1119, various sensors 111 to 118, a storage unit 180, and a control unit 190. The sensor device 10 may also have a location information sensor to automatically detect the location where it is installed. The location information sensor is, for example, a GPS (Global Positioning System) module. Alternatively, the location information sensor may detect the current location of the sensor device 10 from the location of the wireless base station to which the sensor device 10 is connected.

[0048] The communication unit 120 performs modulation and demodulation processing for the sensor device 10 to communicate with other devices. The communication unit 120 performs transmission processing on the signal generated by the control unit 190 and transmits it to an external source (for example, the server 20). The communication unit 120 performs reception processing on the signal received from the external source and outputs it to the control unit 190.

[0049] The touch-sensitive device 11191 provided on the touch panel 1119 is an example of an input device. The input device may be implemented by, for example, a keyboard, a mouse, etc. The input device may also be implemented by, for example, a microphone to respond to voice input from the user. The microphone receives voice input and provides an audio signal corresponding to the voice input to the control unit 190. The input device may include, for example, a receiving port that receives electrical signals input from an external input device.

[0050] The display 11192 provided on the touch panel 1119 is an example of an output device. The display 11192 displays data corresponding to the control of the control unit 190. The output device is a device for presenting information to the user operating the sensor device 10. The output device may be implemented, for example, by a speaker to support audio output to the user. The speaker converts the audio signal provided by the control unit 190 into sound and outputs the sound to the outside of the sensor device 10.

[0051] The memory unit 180 stores data and programs used by the sensor device 10. For example, the memory unit 180 stores calibration information 181, measurement information 182, other sensor measurement information 183, calibration measurement information 184, and learned models 185.

[0052] Calibration information 181 stores information regarding the calibration of the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, NO3 sensor 115, and TUR sensor 117. Calibration information 181 stores, for example, information regarding the calibration of the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, NO3 sensor 115, or TUR sensor 117 transmitted from the server 20. Calibration information 181 may also store information obtained as a result of calibration performed on, for example, the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, NO3 sensor 115, or TUR sensor 117. Calibration information 181 may also read and store calibration information stored in, for example, the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, NO3 sensor 115, or TUR sensor 117. In this embodiment, the calibration information includes, for example, a reference value used when calculating a numerical value, a correction value used when calculating a numerical value, and so on.

[0053] The measurement information 182 stores information obtained by measurement. This information includes, for example, first measurement data, second measurement data, and third measurement data.

[0054] The first measurement data represents data measured by a measurement mechanism such as an EC sensor 111, an FCL sensor 112, a pH sensor 113, an ORP sensor 114, a NO3 sensor 115, a FLOW sensor 116, a TUR sensor 117, or a TEMP sensor 118.

[0055] The second measurement data represents, for example, the first measurement data measured by each measurement mechanism, corrected by a trained model stored within the sensor.

[0056] Specifically, for example, the first measurement data measured by the measurement mechanism of the EC sensor 111 is corrected by a trained model stored in the EC sensor 111 to become the second measurement data. Also, for example, the first measurement data measured by the measurement mechanism of the FCL sensor 112 is corrected by a trained model stored in the FCL sensor 112 to become the second measurement data. Also, for example, the first measurement data measured by the measurement mechanism of the pH sensor 113 is corrected by a trained model stored in the pH sensor 113 to become the second measurement data. Also, for example, the first measurement data measured by the measurement mechanism of the ORP sensor 114 is corrected by a trained model stored in the ORP sensor 114 to become the second measurement data. Also, for example, the first measurement data measured by the measurement mechanism of the NO3 sensor 115 is corrected by a trained model stored in the NO3 sensor 115 to become the second measurement data. Furthermore, for example, the first measurement data measured by the measurement mechanism of the FLOW sensor 116 is corrected by a trained model stored in the FLOW sensor 116 to become the second measurement data. Also, for example, the first measurement data measured by the measurement mechanism of the TUR sensor 117 is corrected by a trained model stored in the TUR sensor 117 to become the second measurement data. Also, for example, the first measurement data measured by the measurement mechanism of the TEMP sensor 118 is corrected by a trained model stored in the TEMP sensor 118 to become the second measurement data.

[0057] The third measurement data represents data calculated from the second measurement data output from each sensor and the calibration information corresponding to each sensor. For example, the third measurement data represents the EC value for the EC sensor 111. The third measurement data represents the FCL value for the FCL sensor 112. The third measurement data represents the pH value for the pH sensor 113. The third measurement data represents the ORP value for the ORP sensor 114. The third measurement data represents the nitric acid concentration for the NO3 sensor 115. The third measurement data represents the turbidity for the TUR sensor 117. For sensors without calibration information, for example, there is no third measurement data.

[0058] Other sensor measurement information 183 stores, for example, information obtained from measurements of other sensor devices 10 that measure water from the same water source. Other sensor devices 10 that measure water from the same water source can be replaced with other sensor devices 10 installed at different locations in the same water treatment facility.

[0059] Calibration measurement information 184 stores information measured when calibration processing is performed on a sensor. Specifically, for example, calibration measurement information 184 stores information measured when calibration processing is performed on the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, NO3 sensor 115, and TUR sensor 117.

[0060] The trained model 185 is a model generated by, for example, having a machine learning model perform machine learning on the server 20 according to a model training program. The trained model 185 is, for example, a parameterized composite function composed of multiple functions that perform predetermined inference based on input data. A parameterized composite function is defined by a combination of multiple tunable functions and parameters. The trained model according to this embodiment may be any parameterized composite function that satisfies the above requirements.

[0061] For example, if the trained model 185 is generated using a feedforward multilayer network, the parameterized composite function is defined as a combination of linear relationships between layers using weight matrices, nonlinear (or linear) relationships using activation functions in each layer, and biases. The weight matrices and biases are called the parameters of the multilayer network. The parameterized composite function changes its form depending on how the parameters are chosen. In a multilayer network, by appropriately setting the constituent parameters, it is possible to define a function that can output desirable results from the output layer.

[0062] As the multilayer network according to this embodiment, for example, a deep neural network (DNN), which is a multilayer neural network targeted by deep learning, may be used. As the DNN, for example, a recurrent neural network (RNN) that targets time-series information may be used.

[0063] The trained model 185 is a model that, for example, when information obtained by measurement is input, outputs whether or not a sensor with an abnormality is included among the sensors that acquired the information. Specifically, the trained model 185 is a model that, for example, when information obtained by measurement is input, outputs whether or not the sensor that measured the information is malfunctioning. The trained model 185 is trained, for example, using multiple measurement values ​​acquired at predetermined intervals by multiple sensors as input data, and the presence or absence of a malfunction in any of the sensors as the correct output data. In other words, the trained model 185 is trained using past measurement information.

[0064] The trained model 185 may be trained, for example, by taking multiple measurement values ​​acquired at predetermined intervals by multiple sensor devices 10 installed in a facility sharing the same water source as input data, and the presence or absence of a malfunction in any of the sensors of the sensor devices 10 as the correct output data. Facilities sharing the same water source refer to, for example, the same water treatment facility.

[0065] When a sensor malfunctions, only the measurement value from that sensor changes drastically. This drastically altered value continues to be measured. According to the trained model 185, it is possible to identify this trend and detect the malfunctioning sensor.

[0066] Furthermore, the trained model 185 is a model that, for example, when information obtained by measurement is input, outputs whether or not there is a deviation from the calibration time in the sensor that measured that information. The trained model 185 is learned by taking multiple measurement values ​​acquired at predetermined intervals by multiple sensors as input data, and whether or not there is a deviation from the calibration time in any of the sensors as the correct output data. In other words, the trained model 185 is learned using past measurement information.

[0067] The trained model 185 may, for example, be trained using multiple measurement values ​​acquired at predetermined intervals by multiple sensor devices 10 installed in a facility sharing the same water source as input data, and the presence or absence of deviations from the calibration time occurring in any of the sensors of the sensor devices 10 as the correct output data.

[0068] When deviations occur from the calibration date, the difference between the measurement value of that sensor and the measurement values ​​of other sensors gradually changes. According to the trained model 185, it is possible to grasp this trend and detect sensors that have deviated from their calibration date.

[0069] The trained model 185 may be trained using a combination of at least one of the above methods.

[0070] The trained model 185 is retrained on server 20, for example, by referencing newly accumulated data.

[0071] The control unit 190 controls the operation of the sensor device 10. For example, the control unit 190 operates according to a program stored in the memory unit 180, performing functions as an operation reception unit 191, a first transmission / reception unit 192, a second transmission / reception unit 193, a calibration unit 194, a calculation unit 195, an estimation unit 196, a correction unit 197, a supplementation unit 198, and a presentation control unit 199.

[0072] The operation reception unit 191 processes instructions or information input from the input device. Specifically, for example, the operation reception unit 191 receives information based on instructions input from the touch-sensitive device 131, etc.

[0073] The first transmitting / receiving unit 192 performs processing to enable the sensor device 10 to send and receive data with an external device such as the server 20 in accordance with a communication protocol. Specifically, for example, the first transmitting / receiving unit 192 transmits at least one of the pieces of information stored in the storage unit 180 to the server 20. Also, for example, the first transmitting / receiving unit 192 receives calibration information from the server 20.

[0074] The second transceiver 193 performs processing for the control unit 190 to send and receive data to and from sensors 111 to 118. Specifically, for example, the second transceiver 193 receives data output from sensors 111 to 118. More specifically, for example, the second transceiver 193 receives first measurement data and second measurement data output from sensors 111 to 118. Also, for example, the second transceiver 193 transmits information received from the server 20 to sensors 111 to 118. The information provided by the server 20 may be, for example, a trained model for each of sensors 111 to 118.

[0075] The calibration unit 194 performs calibration processing for each sensor. For example, the calibration unit 194 performs calibration of sensors 111 to 115 that share the flow cell 1111 all at once. The combined calibration of sensors 111 to 115 is performed, for example, by filling the flow cell 1111 with three different calibration solutions. The first calibration solution is, for example, ultrapure water to determine the baseline. The second calibration solution is, for example, a liquid containing components of the optical system. The third calibration solution is, for example, a liquid containing components that cause chemical effects. The calibration unit 194 stores the calibration information obtained from the combined calibration of sensors 111 to 115 in the calibration information 181.

[0076] The calibration unit 194 estimates whether the calibration process for sensors 111 to 115 was successful based on the changes in the measured values. In the calibration process, if the measured value remains within a predetermined range for a set period of time during measurement using a calibration solution, the calibration process is deemed successful. However, this specification takes time from the start to the end of the calibration. In this embodiment, the calibration unit 194 stores the changes in the measured value for each sensor until it falls within a predetermined range, based on past measurement records. The calibration unit 194 estimates whether the calibration process was successful by comparing the stored changes with the changes in the measured values. The calibration unit 194 may also use a trained model that is configured to output whether the calibration process was successful or not when given the changes in the measured values ​​as input.

[0077] The calibration unit 194 performs calibration of the TUR sensor 117. The calibration unit 194 estimates, for example, whether the calibration process of the TUR sensor 117 was successful based on the trend of the measured values. The calibration unit 194 stores the calibration information obtained from the calibration of the TUR 117 in the calibration information 181.

[0078] Furthermore, if calibration information is provided by the server 20, or if calibration information is read from sensors 111-115 and 117, the calibration unit 194 does not need to perform the calibration process. The calibration unit 194 may, for example, determine whether calibration is required based on information obtained from measurements taken by the sensor device 10. For example, if a sensor's measured values ​​change in a different way than other sensors over time, the calibration unit 194 may determine that calibration is required for that sensor. The calibration unit 194 may also determine that calibration is required after a predetermined period of time has elapsed. If the calibration unit 194 determines that calibration is required, it automatically performs the calibration process.

[0079] The calculation unit 195 performs a process to calculate third measurement data from the second measurement data output from each sensor and the calibration information corresponding to each sensor.

[0080] The estimation unit 196 performs a process to estimate whether or not an abnormality has occurred in sensors 111 to 118. For example, the estimation unit 196 inputs the information obtained from measurements of sensors 111 to 118 into a trained model 185 and outputs whether or not there is an abnormality in the sensor that detected the measured information. Specifically, for example, the estimation unit 196 inputs the information obtained from measurements of sensors 111 to 118 into a trained model 185 and outputs whether or not a faulty sensor is included in the sensor that detected the measured information, and whether or not a sensor that has deviated from its calibration state is included.

[0081] The estimation unit 196 may present the estimation result to the user via the presentation control unit 199.

[0082] If the correction unit 197 detects a sensor that has deviated from the calibration time, it performs a process to correct the measurement data from that sensor using measurement data from multiple sensors acquired in the past and measurement data from other sensors measured in the current measurement.

[0083] If a sensor has deviated from its calibration time, the correction unit 197 may correct the measurement data from that sensor using measurement data acquired in the past by multiple sensor devices 10 installed in facilities sharing the same water source, along with the measurement data taken this time.

[0084] If the interpolation unit 198 includes a sensor that has malfunctioned, it performs a process to interpolate the measurement data from that sensor using measurement data from multiple sensors acquired in the past and measurement data from other sensors measured in the current measurement.

[0085] If a sensor that has malfunctioned is included, the supplementary unit 198 may supplement the measurement data from that sensor using measurement data acquired in the past by multiple sensor devices 10 installed in facilities sharing the same water source, along with the measurement data acquired this time.

[0086] The display control unit 199 presents information related to the processing in the control unit 190 to the user. Specifically, the display control unit 199 displays the information related to the processing in the control unit 190 on the touch panel 1119. The display control unit 199 also presents the information related to the processing in the control unit 190 to the terminal device 30 via the communication unit 120. For example, it displays it on the display unit of the terminal device 30. It also outputs sound to the speaker of the terminal device 30.

[0087] <1.2 Sensor Configuration> The configurations of sensors 111-115, 117, and 118 will be described below. The configuration of pH sensor 113 will be used as an example, but the other sensors also have a measurement mechanism and a circuit board, and have a similar configuration to pH sensor 113.

[0088] Figure 7 is a schematic diagram showing the configuration of the pH sensor 113. The pH sensor 113 shown in Figure 7 has a pH electrode 1131 and a substrate 1132. The pH electrode 1131 is an example of the measurement mechanism of the pH sensor 113. The measurement mechanism may include an element for detecting temperature.

[0089] The circuit board 1132 is equipped with a CPU 11321, memory 11322, amplifier 11323, A / D converter 11324, input / output IF 11325, and communication unit 11326.

[0090] Memory 11322 is, for example, a non-volatile memory. Memory 11322 stores the algorithm used to measure water quality with the pH electrode 1131, the trained model 113221, and the measurement results.

[0091] The trained model 113221 is a model generated by, for example, having a machine learning model perform machine learning on the server 20, according to a model learning program. The trained model 113221 is a model that, when information obtained by measurement is input, reduces the noise contained in the input information and outputs it. The trained model 113221 is learned, for example, by taking as input data a series of measurements that are continuous over time and have at least some discontinuous values, and by using the correct values ​​of the discontinuous measurements contained in the input data as ground truth output data. The trained model 113221 may also be trained using past measurement information.

[0092] The trained model 113221 is retrained on the server 20, for example, based on newly accumulated data. The retrained trained model 113221 is transmitted to the sensor device 10 and replaces the trained model 113221 stored in memory 11322.

[0093] Memory 11322 may store information regarding the calibration of the pH sensor 113. For example, the pH sensor 113 may be shipped with the learned model 113221 and calibration information stored in memory 11322. The calibration information stored in memory 11322 is read from memory 11322 when the pH sensor 113 is connected to the control box 1112, and stored in the storage unit 180 of the sensor device 10.

[0094] Amplifier 11323 amplifies the analog signal measured by pH electrode 1131.

[0095] The A / D converter 11324 converts the amplified analog signal into a digital signal. The digital signal represents the first measurement data described above.

[0096] The CPU 11321 comprehensively controls the operation of the pH sensor 113. For example, the CPU 11321 stores the first measurement data in memory 11322. The CPU 11321 corrects any anomalies contained in the first measurement data. In other words, the CPU 11321 reduces the noise contained in the first measurement data. Specifically, for example, the CPU 11321 inputs the first measurement data into a trained model 113221 to replace measurement values ​​that have changed abruptly with values ​​estimated from previous measurement values. The data corrected by the trained model 113221 represents the second measurement data. For example, the CPU 11321 stores the second measurement data in memory 11322.

[0097] The input / output IF11325 is an interface for connecting the sensor device 10 to the control box 1112. The input / output IF11325 transmits, for example, the first measurement data and the second measurement data to the control box 1112. If calibration information is stored in the memory 11322, the input / output IF11325 transmits calibration information to the control box 1112. The input / output IF11325 receives information output from the control box 1112, for example, a newly updated trained model.

[0098] The communication unit 11326 performs processing such as modulation / demodulation processing for the pH sensor 113 to communicate with other devices. The communication unit 11326 receives information transmitted from the server 20, for example, a newly updated trained model.

[0099] <1.3 Server Configuration> Figure 8 shows an example of the functional configuration of server 20. As shown in Figure 8, server 20 functions as a communication unit 201, a storage unit 202, and a control unit 203.

[0100] The communications unit 201 performs processing to enable the server 20 to communicate with external devices.

[0101] The memory unit 202 includes, for example, a plant table 2021, an installation table 2022, a plant environment table 2023, a measurement table 2024, a calibration table 2025, a model table 2026, a first trained model 2027, and a second trained model 2028.

[0102] Plant Table 2021 is a table that stores information about water treatment facilities. Further details will be provided later.

[0103] Installation table 2022 is a table that stores the installation locations of the sensor device 10. Further details will be described later.

[0104] The Plant Environment Table 2023 is a table that stores environmental information for the area where a water treatment facility will be established. Further details will be provided later.

[0105] Measurement table 2024 is a table that stores data measured by the sensor device 10. Further details will be described later.

[0106] Calibration table 2025 is a table that stores data related to calibration. Further details will be provided later.

[0107] Model table 2026 is a table that manages the versions of the first trained model 2027 and the second trained model 2028. Further details will be provided later.

[0108] The first pre-trained model 2027 and the second pre-trained model 2028 are models generated by, for example, having a machine learning model perform machine learning on server 20, according to a model training program. The first pre-trained model 2027 and the second pre-trained model 2028 are models that, when data is input, are trained to output predetermined information based on the input data. The first pre-trained model 2027 and the second pre-trained model 2028 are separate pre-trained models that are trained with different training data, for example, depending on the information to be output.

[0109] The first pre-trained model 2027 is a model that, for example, takes information about a given water treatment facility as input and outputs an analysis result of the water treatment at that facility. Specifically, the first pre-trained model 2027, for example, takes information about a given water treatment facility as input and outputs an evaluation of the water treatment at that facility. The output as an evaluation may be in alphabetical order such as A, B, C, or it may be a score with a maximum of 100 points. The first pre-trained model 2027 is trained, for example, using information about multiple water treatment facilities as input data and evaluations of water treatment facilities as ground truth output data.

[0110] Multiple types of the first pre-trained model 2027 may be created depending on the criteria for analysis. For example, when evaluating water treatment based on water usage, the first pre-trained model 2027 is trained to use evaluations of water usage as ground truth output data. Alternatively, when evaluating water treatment based on pollutant emissions, the first pre-trained model 2027 is trained to use evaluations of pollutant emissions as ground truth output data. Similarly, when evaluating water treatment based on water reuse, the first pre-trained model 2027 is trained to use evaluations of water reuse as ground truth output data. Furthermore, when evaluating water treatment comprehensively, such as with an environmental consideration score, the first pre-trained model 2027 is trained to use comprehensive evaluations that reference multiple pieces of information about water treatment facilities as ground truth output data.

[0111] Furthermore, the first pre-trained model 2027 may have multiple variations depending on classifications that have similar water quality levels required for water treatment. For example, when water treatment facilities are installed in hospitals, train stations / airports, factories, hotels, and simple water supply systems, the water quality levels required for water treatment differ at each facility. Therefore, water treatment facilities may be classified according to their installation location or the type of facility they are installed in. If classifications are set for water treatment facilities, the first pre-trained model 2027 is trained using information on multiple water treatment facilities with the same installation location classification as input data, and evaluations of the water treatment facilities as ground truth output data.

[0112] Information regarding water treatment facilities includes, for example, the following: ·Raw water quality • Construction area • Plant type • Processing equipment and processing details provided • Operating status of the processing unit • Water usage • Emissions of pollutants • Amount of water reuse • Chemicals used • Environmental information (weather, temperature, humidity, wind speed, atmospheric pressure, dust) • Measurement data from various sensors

[0113] The second pre-trained model 2028 is a model that, for example, estimates the water treatment at a water treatment facility when given design data and water quality data for that facility. Specifically, the second pre-trained model 2028 outputs what kind of water treatment will be performed at the designed water treatment facility when given design data and water quality data for a water treatment facility. The second pre-trained model 2028 is trained, for example, by taking information on multiple water treatment facilities as input data and similar water treatment facilities as ground truth output data.

[0114] The design data for the water treatment facility input into the second trained model 2028 includes, for example, the following: • Construction area • Plant type • Processing equipment to be provided, processing details • Piping structure of water treatment facilities • Location and number of processing units to be installed. • Unit treatment water volume • Chemicals used

[0115] The first trained model 2027 and the second trained model 2028 are retrained on server 20, for example, based on newly accumulated data.

[0116] The control unit 203 operates according to the program stored in the memory unit 202, thereby performing the functions shown as the reception control module 2031, transmission control module 2032, memory control module 2033, analysis module 2034, simulation module 2035, calibration setting module 2036, proposal module 2037, learning module 2038, and presentation module 2039.

[0117] The receive control module 2031 controls the process by which the server 20 receives signals from external devices according to a communication protocol.

[0118] The transmission control module 2032 controls the process by which the server 20 transmits signals to external devices according to a communication protocol. For example, the transmission control module 2032 transmits a trained model to the sensor device 10.

[0119] The memory control module 2033 controls the process of storing received data in various tables of the memory unit 202. For example, when information about a predetermined water treatment facility is entered into the server 20 at the time of a service contract, the memory control module 2033 stores the entered information in the plant table 2021. When the memory control module 2033 receives information that the equipment of the water treatment facility has been changed, it updates the information stored in the plant table 2021.

[0120] The control unit 203 receives information about the mounting location of the sensor device 10 in the water treatment facility. When the memory control module 2033 receives information about the mounting location of the sensor device 10, it stores the input information in the installation table 2022.

[0121] The control unit 203 receives information about the surrounding environment where the water treatment facility is located. When the memory control module 2033 receives information about the surrounding environment, it stores the input information in the plant environment table 2023.

[0122] The control unit 203 receives measurement data measured by the sensor device 10. When the memory control module 2033 receives measurement data measured by the sensor device 10, it stores the received information in the measurement table 2024.

[0123] When calibration information is set by the calibration setting module 2036, the memory control module 2033 stores the set information in the calibration table 2025, associating it with the setting date.

[0124] When the training module 2038 retrains a trained model, the memory control module 2033 assigns a model ID to the retrained model and stores the model ID, along with version information and creation date, in the model table 2026.

[0125] The memory control module 2033, for example, forms a memory area for each water treatment facility. The memory control module 2033 may set the capacity of the memory area based on the terms of the contract the water treatment facility has. For example, the memory control module 2033 allocates a maximum of the first capacity to water treatment facilities with a free contract. For example, the memory control module 2033 allocates a maximum of the second capacity, which is larger than the first capacity, to water treatment facilities with a paid contract. For example, the memory control module 2033 allocates an unlimited amount of memory area capacity to water treatment facilities with a premium contract.

[0126] The memory control module 2033 may restrict the measurement items that can be stored depending on the contract it has. For example, the memory control module 2033 may prevent the storage of data measured by the NO3 sensor 115 for water treatment facilities with a free contract. The memory control module 2033 may also restrict the period for which data can be stored depending on the contract it has. For example, the memory control module 2033 may prevent the storage of data for water treatment facilities with a free contract from being stored for a shorter period than for water treatment facilities with a paid contract. The memory control module 2033 may also restrict the sampling period for data storage depending on the contract it has. For example, the memory control module 2033 may sample data at a longer period for water treatment facilities with a free contract than for water treatment facilities with a paid contract.

[0127] The analysis module 2034 controls the process of analyzing water treatment in a water treatment facility. Specifically, for example, when a predetermined water treatment facility is specified, the analysis module 2034 evaluates the water treatment in the specified facility based on information about the specified facility. This evaluation is, for example, an example of analysis. More specifically, for example, the analysis module 2034 inputs information about the specified water treatment facility into the first trained model 2027, and the first trained model 2027 outputs an evaluation of the water treatment.

[0128] The analysis module 2034 may allow the user to select criteria for water treatment analysis. When the user specifies the criteria for analysis, the analysis module 2034 selects a first pre-trained model 2027 of the type corresponding to the specified criteria, inputs information about the water treatment facility into the selected first pre-trained model 2027, and has the first pre-trained model 2027 output an evaluation.

[0129] The analysis module 2034 may accept a selection from the user regarding the classification of the location of the water treatment facility. When a classification is specified by the user, the analysis module 2034 selects a first pre-trained model 2027 of the type corresponding to the specified classification, inputs information about the water treatment facility into the selected first pre-trained model 2027, and has the first pre-trained model 2027 output an evaluation.

[0130] The analysis module 2034 may set the version of the first pre-trained model 2027 to be used based on the terms of the contract the water treatment facility has. For example, for water treatment facilities with a free contract, the analysis module 2034 will analyze the water treatment using a version of the first pre-trained model 2027 that is not the latest version. A version of the first pre-trained model 2027 that is not the latest version refers to, for example, a version of the first pre-trained model 2027 that is several generations older than the latest version that is open for use. Alternatively, for water treatment facilities with a paid or premium contract, the analysis module 2034 will analyze the water treatment using the latest version of the first pre-trained model 2027.

[0131] Furthermore, the analysis module 2034, for example, when a predetermined water treatment facility is specified, estimates the amount of power consumed at the specified water treatment facility based on information about that facility. The analysis module 2034 may also estimate the amount of power consumed using a trained model.

[0132] Simulation module 2035 controls the process of estimating water treatment at a hypothetical water treatment facility. Estimation is, for example, an example of analysis. Specifically, when simulation module 2035 receives, for example, design data of a hypothetical water treatment facility and water quality data of the water treated at this facility as input, it estimates the water treatment at the hypothetical water treatment facility based on the input design data and water quality data. More specifically, for example, simulation module 2035 inputs the design data of the hypothetical water treatment facility and the water quality data of the water treated at this facility into a second pre-trained model 2028, and outputs information from the second pre-trained model 2028 that estimates the water treatment at the hypothetical water treatment facility. Simulation module 2035 may, for example, construct a model of the water treatment facility based on the input design data and water quality data, and then simulate the water treatment at the hypothetical water treatment facility.

[0133] Simulation module 2035 may set the version of the second pre-trained model 2028 to be used based on the terms of the contract the water treatment facility has. For example, for water treatment facilities with a free contract, simulation module 2035 will simulate water treatment using a version of the second pre-trained model 2028 that is not the latest version. A version of the second pre-trained model 2028 that is not the latest version refers to, for example, a version of the second pre-trained model 2028 that is several generations older than the latest version that is open for use. Also, for example, for water treatment facilities with a paid or premium contract, simulation module 2035 will analyze water treatment using the latest version of the second pre-trained model 2028.

[0134] The calibration setting module 2036 controls the process of calibrating the sensors of the sensor device 10. Specifically, for example, the calibration setting module 2036 sets information regarding the calibration of the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, NO3 sensor 115, and TUR sensor 117 of the sensor device 10.

[0135] The calibration setting module 2036 determines, for example, whether calibration is required based on information obtained from measurements taken by the sensor device 10. For example, if a sensor's measured values ​​change over time in a different trend than other sensors, the calibration setting module 2036 determines that the sensor needs calibration. The calibration setting module 2036 may also determine that calibration is required after a predetermined period of time has elapsed.

[0136] The calibration setting module 2036 calculates calibration information for a sensor if there is a sensor that requires calibration. The calibration setting module 2036 may calculate calibration information in response to a request from the sensor device 10. Specifically, the calibration setting module 2036 calculates calibration information from, for example, measurement information from the sensor device 10 stored in the memory unit 202 and calibration information set in the past. The calibration setting module 2036 may calculate calibration information using, for example, a trained model, or it may grasp a predetermined trend and calculate calibration information using the grasped trend. In this case, the trained model is learned, for example, using measurement information from the sensor device 10 measured in the past as input data and calibration information set in the past as the correct output data.

[0137] The proposal module 2037 controls the process of proposing improvement plans for water treatment facilities. Specifically, for example, the proposal module 2037 proposes improvement plans to the user, which are output by the analysis module 2034, to improve the evaluation of the water treatment at a given water treatment facility. Improvement plans to improve the evaluation of water treatment include, for example, the following: • Proposal of new items to be sensed • Proposal of new locations to be sensed • Proposal for operational control

[0138] A proposal for new items to be sensed means, for example, suggesting new items to be measured among the items related to water. Similarly, a proposal for new locations to be sensed means, for example, suggesting new locations to measure items related to water. Furthermore, a proposal for operational control means, for example, suggesting changes to the control system that operates the treatment equipment installed in a water treatment facility. Operational control proposals include, for example, adjusting thresholds and changing trigger signals.

[0139] The proposed module 2037 may, for example, set improvement plans using a pre-trained model, or it may set improvement plans based on the structure of similar water treatment facilities. In this case, the pre-trained model is trained, for example, using information about water treatment facilities as input data and the above-mentioned improvement plans proposed for a given water treatment facility as ground truth output data.

[0140] Proposal module 2037 may provide the user with an estimate for implementing the proposed improvements. Alternatively, proposal module 2037 may also estimate the operating costs and provide the user with these estimated costs.

[0141] Proposed module 2037 may propose appropriate measurement data trends as improvement targets. For example, proposed module 2037 may set improvement targets using a trained model, or it may set improvement targets based on the measurement data trends of water treatment facilities with high water treatment performance ratings among similar water treatment facilities.

[0142] Furthermore, for example, the proposal module 2037 suggests to the user improvements to enhance the water treatment estimation results output by the simulation module 2035. These improvements to enhance the water treatment estimation results include, for example, the following: • Proposal of new items to be sensed • Proposal of new locations to be sensed • Proposal for operational control

[0143] Furthermore, for example, the suggestion module 2037 proposes measures to the user to reduce the power consumption estimated by the analysis module 2034. These measures to reduce the estimated power consumption include, for example, the following: • Stop the equipment (pumps, blowers, turbines, etc.) that is driving the processing unit. • Reducing the number of devices that drive the processor. • Reduce the output of the equipment driving the processor. If the equipment related to the proposed countermeasures can be directly controlled from the server 20, the countermeasures may be automatically applied to the equipment.

[0144] The learning module 2038 controls the process of generating trained models. Specifically, the learning module 2038 generates trained models by, for example, having the machine learning model perform machine learning according to the model learning program. More specifically, for example, the learning module 2038 generates a first trained model 2027 or a second trained model 2028. Alternatively, for example, the learning module 2038 generates a trained model 185 or a trained model 113221 and transmits it to the sensor device 10 via the communication unit 201.

[0145] Furthermore, the learning module 2038 retrains the trained model at predetermined intervals. Specifically, for example, the learning module 2038 retrains the first trained model 2027 or the second trained model 2028. When the learning module 2038 retrains the first trained model 2027 or the second trained model 2028, it stores the trained model generated by the retraining in the storage unit 202 as a separate model from the model before retraining. Also, for example, the learning module 2038 retrains the trained model 185 or the trained model 113221. The learning module 2038 transmits the retrained trained model 185 or the trained model 113221 to the sensor device 10 via the communication unit 201.

[0146] The presentation module 2039 controls the process of presenting data managed by the server 20 to the user. Specifically, for example, the presentation module 2039 presents the analysis results created by the analysis module 2034 to a user who has specified a particular water treatment facility. The presentation module 2039 also presents the simulation results created by the simulation module 2035 to a user who has requested a simulation of a hypothetical water treatment facility.

[0147] Furthermore, the presentation module 2039 presents the user with a warning (alert) based on the power consumption estimated by the analysis module 2034. The alert may be displayed as an image or output as audio. Along with the alert, the presentation module 2039 also estimates and presents to the user the increase in operational costs that may result from not responding to the alert.

[0148] Presentation module 2039 presents the requested information to a user when the user, who is associated with a given water treatment facility, requests at least one piece of information related to that water treatment facility. Presentation module 2039 presents the requested information in any manner desired by the user. Desired manners include, for example, a manner that utilizes predetermined statistical rules that enable the discovery of data features. This makes it easier for users to utilize the data by allowing them to view any data in any manner they choose.

[0149] Presentation module 2039 may set the acceptable data presentation methods based on the terms of the contract the water treatment facility has with it. For example, presentation module 2039 may restrict the available tools for water treatment facilities with a free service contract.

[0150] <2 Data Structure> Figures 9 to 14 show the data structure of tables stored by server 20. Note that Figures 9 to 14 are examples and do not exclude data not shown. Furthermore, even data listed in the same table may be stored in separate memory areas within the storage unit 202.

[0151] Figure 9 shows the data structure of Plant Table 2021. Plant Table 2021, shown in Figure 9, is a table with Plant ID as the key and columns for Plant Name, Address, Contract Details, Raw Water Quality, Type, Treatment Equipment, Treatment Details, and Chemicals Used. Plant Table 2021 may also have a column for classification assigned to the location where the water treatment facility is installed.

[0152] The Plant ID is an item that stores an identifier to uniquely identify a water treatment facility (water treatment plant). The Plant Name is an item that stores the name of the water treatment facility. The Address is an item that stores the address where the water treatment facility is located. The Contract Details is an item that shows the content of the contract that the water treatment facility has entered into, such as free, paid, or premium membership. The Contract Details also stores the date on which the paid contract was terminated, i.e., the date on which it became a free contract. The Raw Water Quality is an item that stores the quality of the water used for water treatment in the area where the water treatment facility is located. The Type is an item that shows the type of water treatment facility. Types include, for example, groundwater utilization facilities, sewage treatment facilities, or water purification facilities. The Treatment Equipment is an item that shows the treatment equipment installed in the water treatment facility, such as raw water tanks and pre-filters. The Treatment Content is an item that shows the content of the treatment carried out in the treatment equipment. The Chemicals Used is an item that shows the names of the chemicals used in the treatment.

[0153] Figure 10 shows the data structure of the installation table 2022. The installation table 2022 shown in Figure 10 is a table that has columns for plant ID, installation area, and installation date, with sensor ID as the key.

[0154] The Sensor ID is an item that stores an identifier to uniquely identify the sensor device 10. The Plant ID is an item that indicates the water treatment facility where the sensor device 10 is installed. The Installation Area is an item that stores the area within the water treatment facility where the sensor device 10 is installed. The area where the sensor device 10 is installed is associated, for example, with the area in the water treatment facility where the treatment equipment is installed. The Installation Date is an item that stores the year, month, and day the sensor device 10 was installed.

[0155] Figure 11 shows the data structure of the plant environment table 2023. The plant environment table 2023 shown in Figure 11 is a table with the plant ID as the key and columns for measurement date, weather, temperature, humidity, wind speed, atmospheric pressure, and dust.

[0156] The measurement date is an item that stores the date on which environmental information around the water treatment facility was measured. The weather is an item that stores the weather on the measurement date. The temperature is an item that stores the temperature on the measurement date. The humidity is an item that stores the humidity on the measurement date. The wind speed is an item that stores the wind speed on the measurement date. The atmospheric pressure is an item that stores the atmospheric pressure on the measurement date. The dust is an item that stores the concentration of dust, such as yellow dust, on the measurement date.

[0157] Figure 12 shows the data structure of measurement table 2024. Measurement table 2024 shown in Figure 12 is a table that has columns for measurement date and time, sensor, first measurement value, second measurement value, and third measurement value, with sensor ID as the key.

[0158] The Sensor ID is an item that stores an identifier to uniquely identify the sensor device 10. The Measurement Date and Time is an item that stores the date and time on which the information obtained by measurement was received from the sensor device 10. Specifically, the Measurement Date and Time is an item that stores the date and time on which the first measurement data, second measurement data, and third measurement data were received from the sensor device 10. The Sensor is an item that indicates the name to identify the various sensors included in the sensor device 10. The First Measurement Value is an item that stores the value of the first measurement data measured by various sensors 111 to 118. The Second Measurement Value is an item that stores the value of the second measurement data obtained by correcting the first measurement data with a trained model stored in, for example, sensors 111 to 115, 117, and 118. The Third Measurement Value is an item that stores the value of the third measurement data calculated from the second measurement data and the calibration information corresponding to each sensor.

[0159] The sensors are labeled, for example, "EC" for the EC sensor 111, "FCL" for the FCL sensor 112, "pH" for the pH sensor 113, "ORP" for the ORP sensor 114, "NO3" for the NO3 sensor 115, "FLOW" for the FLOW sensor 116, "TUR" for the TUR sensor 117, and "Temp" for the TEMP sensor 118.

[0160] Figure 13 shows the data structure of calibration table 2025. Calibration table 2025, shown in Figure 13, is a table with sensor ID as the key and columns for sensor, setup date, and calibration information.

[0161] The Sensor ID is an item that stores an identifier to uniquely identify the sensor device 10. The Sensor is an item that indicates the name to identify the various sensors included in the sensor device 10. The Setting Date is an item that stores the date on which the calibration information was set. The Calibration Information is an item that stores information related to the calibration.

[0162] Figure 14 shows the data structure of Model Table 2026. Model Table 2026, shown in Figure 14, is a table with Model ID as the key and columns for Version Information, Update Date, and Release.

[0163] The Model ID is an identifier that uniquely identifies the trained model. The Version Information is an identifier that stores the version of the trained model. The Update Date is an identifier that stores the date the trained model represented by the Model ID was created. The Release status indicates whether the trained model is released or not. It is True if it is released and False if it is not.

[0164] <3 operations> The operation of the sensor device 10 and server 20 installed in System 1 will be described.

[0165] <3.1 Operation of Sensor Device 10) First, the user of the sensor device 10 prepares for operation. Specifically, the user checks that the wiring and piping of the sensor device 10 are correct. The user turns on the breaker switch located in the control box 1112. The user confirms that the air vent valve 1115 is closed. The user opens valves 119 and 1110 and introduces the sample water into the sensor device 10. As a result, the sample water is supplied from valve 119, and the flow cell 1111 and shell 1171 are filled with the sample water.

[0166] The user checks the flow rate measured by the FLOW sensor 116 on the touch panel 1119 and turns the flow rate adjustment knob on the flow rate adjustment valve 1117. By turning the flow rate adjustment knob, the user adjusts the flow rate of the sample water.

[0167] The user inserts an air venting tube into the inlet 11151 located at the top of the air venting valve 1115. The user opens the air venting valve 1115 to release the air. The measurement ports 11111, 11112, and 11113 of the flow cell 1111 are formed so that their openings face upward, and the air venting passage 11115 is provided near the openings, making it possible to release the air from the measurement ports 11111, 11112, and 11113 all at once. After releasing the air, the user closes the air venting valve 1115.

[0168] Calibration information for the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, NO3 sensor 115, and TUR sensor 117 attached to the sensor device 10 is stored, for example, in the memory unit 180. The sensor device 10 uses the information stored in the memory unit 180 to measure the sample water.

[0169] Calibration information for the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, NO3 sensor 115, and TUR sensor 117 may be pre-written in the memory of each sensor. The control unit 190 receives the calibration information stored in each sensor via the second transmitting / receiving unit 193. The control unit 190 stores the received information in the storage unit 180. When replacing a sensor, the calibration information stored in each sensor may also be received via the second transmitting / receiving unit 193.

[0170] If calibration information is provided by the server 20, the storage unit 180 of the sensor device 10 and the memory of each sensor do not need to store calibration information. The user inputs an instruction to the sensor device 10 to query the server 20 for calibration information.

[0171] (Obtaining information regarding calibration) Figure 15 illustrates an example of the operation in which the sensor device 10 shown in Figure 1 acquires calibration information from the server 20.

[0172] The first transmitting / receiving unit 192 of the control unit 190 queries the server 20 for calibration information (step S11). Specifically, the first transmitting / receiving unit 192 queries the server 20 to find out whether there is any calibration information for, for example, the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, NO3 sensor 115, or TUR sensor 117.

[0173] The control unit 203 of the server 20, using the calibration setting module 2036, checks whether there is any information to send to the sensor device 10 that has been inquired about (step S12). For example, the storage unit 202 of the server 20 stores calibration information for the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, NO3 sensor 115, and TUR sensor 117 of the sensor device 10. The calibration setting module 2036 determines that there is information to send to the sensor device 10 and reads the calibration information from the calibration table 2025 in the storage unit 202 (step S13). The server 20, using the transmission control module 2032, sends the read information to the sensor device 10 (step S14).

[0174] The first transmitting / receiving unit 192 of the control unit 190 receives information transmitted from the server 20 and stores the received calibration information in the calibration information 181 of the storage unit 180 (step S15).

[0175] Furthermore, the provision of calibration information is not limited to providing it in response to a request from the sensor device 10. If the calibration setting module 2036 determines that a specific sensor in the sensor device 10 is in a state where calibration is required, it may transmit calibration information for that sensor to the sensor device 10.

[0176] Furthermore, there may be cases where no new calibration information has been set in response to an inquiry from the sensor device 10. In this case, the calibration setting module 2036 returns to the sensor device 10 that there is no information to send to the sensor device 10. If the sensor device 10 requests calibration in response to the response that there is no information to send, the calibration setting module 2036 calculates the calibration information. The explanation of Figure 15 describes the operation of the sensor device 10 when it is started up. The operation of acquiring calibration information from the server 20 is not limited to being performed when the sensor device 10 is started up. The first transmitting / receiving unit 192 may also acquire calibration information when the sensor is replaced.

[0177] (Calibration process in sensor device 10) The calibration process may be performed by the sensor device 10.

[0178] Figure 16 is a flowchart illustrating an example of the operation in which the sensor device 10 shown in Figures 2 and 3 performs calibration processing.

[0179] When performing the calibration process, the user first closes valves 119 and 1110 to stop the inflow of sample water. The user then attaches a drain tube to the drain valve 1116. The user then opens the drain valve 1116 to discharge the sample water to the outside of the sensor device 10. Note that the calibration process for the sensor device 10 may be performed before supplying sample water to the sensor device 10.

[0180] The user removes sensors 111-115 and 118 from flow cell 1111, wipes off any water droplets, and reinserts them into flow cell 1111. The user then supplies the first calibration solution to flow cells 1111-1 and 1111-2 from which the sample water has been discharged. Once flow cells 1111-1 and 1111-2 are filled with the first calibration solution, the user opens the air release valve 1115 to remove air from flow cells 1111-1 and 1111-2.

[0181] The sensor device 10 displays a "Calibration Start" object on the touch panel 1119 to perform the calibration process. The user presses the "Calibration Start" object displayed on the touch panel 1119. When the "Calibration Start" object is pressed, the sensor device 10 accepts the selection of the sensor to be calibrated (step S21). For example, the sensor device 10 displays a sensor selection screen on the touch panel 1119 and accepts the sensor selection from the user.

[0182] The user selects the sensors to be calibrated simultaneously from among the multiple sensors displayed on the selection screen. For example, the user selects the EC sensor 111, FCL sensor 112, pH sensor 113, ORP sensor 114, and NO3 sensor 115.

[0183] When a sensor is selected, the sensor device 10 accepts the selection of a calibration solution (step S22). For example, the sensor device 10 displays a calibration solution selection screen on the touch panel 1119 and accepts the user's selection of a calibration solution. The user selects the first calibration solution from among the multiple calibration solutions displayed on the selection screen.

[0184] When a calibration solution is selected, the sensor device 10 receives an instruction to start the calibration process (step S23). For example, the sensor device 10 displays a "Start" object on the touch panel 1119 and receives a start instruction from the user. The user presses the "Start" object to start the calibration process.

[0185] When the calibration process is started, the sensor device 10 receives first measurement data from the selected sensor via the second transmitting / receiving unit 193 (step S24). The sensor device 10 stores the received first measurement data in the calibration measurement information 184 of the storage unit 180.

[0186] The sensor device 10 uses a calibration unit 194 to estimate whether the calibration process for sensors 111 to 115 was successful (step S25). Specifically, for example, the calibration unit 194 monitors the changes in the measured values. The calibration unit 194 compares the changes in the measured values ​​obtained with the changes in the measured values ​​that are expected to occur when calibration is successful, which are stored for each sensor. For example, if the changes in the obtained measured values ​​fall within a predetermined range of the expected changes, the calibration unit 194 determines that the calibration was successful.

[0187] If the calibration is successful, the calibration unit 194 determines whether the measurement using the expected calibration solution has been completed (step S26). If the measurement using the expected calibration solution has not been completed (No. in step S26), the calibration unit 194 proceeds to step S22.

[0188] The user, for example, discharges the first calibration solution from flow cells 1111-1 and 1111-2 and supplies the second calibration solution to flow cells 1111-1 and 1111-2. The calibration unit 194 performs the sensor calibration process in flow cells 1111-1 and 1111-2, which are filled with the second calibration solution. If the calibration process using the second calibration solution is successful, the calibration unit 194 moves the process to step S22.

[0189] The user, for example, discharges the second calibration solution from flow cells 1111-1 and 1111-2 and supplies the third calibration solution to flow cells 1111-1 and 1111-2. The calibration unit 194 performs the sensor calibration process in flow cells 1111-1 and 1111-2, which are filled with the third calibration solution. If the calibration process using the third calibration solution is successful, the calibration unit 194 moves the process to step S27, for example.

[0190] If the measurement using the expected calibration solution is completed (Yes in step S26), the calibration unit 194 creates calibration information for each sensor based on the information measured for each sensor (step S27). The calibration unit 194 stores the created calibration information in the calibration information 181 of the storage unit 180.

[0191] The sensor device 10 transmits calibration information stored in calibration information 181 to the server 20 at a predetermined timing using the first transmitting / receiving unit 192. The sensor device 10 also transmits information stored in calibration measurement information 184 to the server 20 at a predetermined timing using the first transmitting / receiving unit 192. The predetermined timing is, for example, as follows: • Predetermined cycle • At the designated time • When the calibration process is complete When information is stored

[0192] (Measurement of water sample: Processing by sensor) The sensor device 10 measures the water quality of the sample water using sensors 111 to 118. Specifically, for example, it measures the water quality of the sample water using sensors 111 to 115 and 118 inserted into the measurement port of the flow cell 1111, and the TUR sensor 117 inserted into the shell 1171.

[0193] Figure 17 is a flowchart illustrating an example of the operation of the sensor shown in Figure 7.

[0194] The pH sensor 113 shown in Figure 7 measures the potential of the sample water filled in the flow cell 1111-1 using the pH electrode 1131, for example, at a preset period (step S31). The pH sensor 113 stores the acquired potential as the first measurement data in the memory 11322.

[0195] The pH sensor 113 corrects any abnormalities in the first measurement data (step S32). Specifically, the pH sensor 113 reduces noise in the first measurement data using, for example, a trained model 113221. Specifically, for example, the pH sensor 113 inputs the first measurement data into the trained model 113221 to replace measurement values ​​that have changed abruptly with values ​​estimated from previous measurement values. Measurement values ​​that have not changed abruptly are output directly from the trained model 113221. The pH sensor 113 stores the data output from the trained model 113221 as second measurement data in memory 11322.

[0196] The pH sensor 113 transmits the first measurement data and the second measurement data to the sensor device 10 via the input / output IF 11325 at predetermined timings. The predetermined timings are, for example, as follows: • Predetermined cycle When information is obtained When information is stored

[0197] Sensors 111-115, 118, and the TUR sensor 117 inserted into shell 1171 operate in the same manner as the pH sensor 113.

[0198] (Measurement of sample water: Processing by sensor device 10) Figure 18 is a flowchart illustrating an example of the operation of the sensor device 10 shown in Figure 6.

[0199] The control unit 190 receives first measurement data and second measurement data from sensors 111-115, 117, and 118 via the second transmitting / receiving unit 193 (step S41). The control unit 190 stores the received first measurement data and second measurement data in the measurement information 182 of the storage unit 180.

[0200] The control unit 190 calculates third measurement data from the second measurement data received from sensors 111-115 and 117 and the calibration information corresponding to each sensor using the calculation unit 195 (step S42).

[0201] Specifically, the calculation unit 195 calculates the EC value as a third measurement data from, for example, the second measurement data received from the EC sensor 111 and the calibration information for the EC sensor 111. The calculation unit 195 also calculates the FCL value as a third measurement data from, for example, the second measurement data received from the FCL sensor 112 and the calibration information for the FCL sensor 112. The calculation unit 195 also calculates the pH value as a third measurement data from, for example, the second measurement data received from the pH sensor 113 and the calibration information for the pH sensor 113. The calculation unit 195 also calculates the ORP value as a third measurement data from, for example, the second measurement data received from the ORP sensor 114 and the calibration information for the ORP sensor 114. The calculation unit 195 also calculates the nitric acid concentration as a third measurement data from, for example, the second measurement data received from the NO3 sensor 115 and the calibration information for the NO3 sensor 115. Furthermore, the calculation unit 195 calculates turbidity as third measurement data from, for example, the second measurement data received from the TUR sensor 117 and information regarding the calibration of the TUR sensor 117.

[0202] The control unit 190 uses the estimation unit 196 to estimate whether or not an abnormality has occurred in sensors 111 to 118 (step S43). For example, the estimation unit 196 inputs the third measurement data calculated for sensors 111 to 115 and 117, the first measurement data measured by the FLOW sensor 116, and the second measurement data measured by the TEMP sensor 118 to the trained model 185. If an abnormality has occurred in any of the sensors from which measurement data has been input, the trained model 185 outputs the sensor that is estimated to have experienced an abnormality and the abnormality estimated to have occurred in that sensor.

[0203] Specifically, if any of the sensors into which measurement data is input are faulty, the trained model 185 will output, for example, the sensor that is estimated to be faulty and that a fault has occurred in that sensor. Also, if any of the sensors into which measurement data is input are deviating from the calibration date, the trained model 185 will output, for example, the sensor that is estimated to be deviating from the calibration date and that a deviation has occurred in that sensor.

[0204] The estimation unit 196 may present the estimation result to the user via the presentation control unit 199.

[0205] The control unit 190 determines, using the interpolation unit 198, whether or not a faulty sensor is included (step S44). If a faulty sensor is included (Yes in step S44), the interpolation unit 198 interpolates the measurement data from the faulty sensor using measurement data from multiple sensors acquired in the past and measurement data from other sensors measured this time (step S45).

[0206] Specifically, for example, if the interpolation unit 198 estimates that a malfunction has occurred in the pH sensor 113, it discards the third measurement data calculated based on the measurement of the pH sensor 113. The interpolation unit 198 uses the third measurement data of sensors 111-115 and 117, the first measurement data of the FLOW sensor 116, and the second measurement data of the TEMP sensor 118, which were acquired in the past, along with the third measurement data of sensors 111, 112, 114, 115, and 117, the first measurement data of the FLOW sensor 116, and the second measurement data of the TEMP sensor 118, which were measured this time, to calculate the interpolation measurement data for the pH sensor 113. The interpolation unit 198 stores the calculated interpolation measurement data as the third measurement data in the storage unit 180.

[0207] If no faulty sensor is included (No. in step S44), the compensation unit 198 proceeds to step S46.

[0208] The control unit 190, using the correction unit 197, determines whether or not a sensor that has deviated from the time of calibration is included (step S46). If a sensor that has deviated from the time of calibration is included (Yes in step S46), the correction unit 197 corrects the measurement data from the sensor that has deviated from the time of calibration using measurement data from multiple sensors acquired in the past and measurement data from other sensors measured this time (step S47).

[0209] Specifically, for example, if the correction unit 197 estimates that a deviation has occurred in the pH sensor 113 since calibration, it corrects the third measurement data calculated based on the measurement of the pH sensor 113. At this time, the correction unit 197 uses the third measurement data of sensors 111-115 and 117, the first measurement data of the FLOW sensor 116, and the second measurement data of the TEMP sensor 118 acquired in the past, along with the third measurement data of sensors 111, 112, 114, 115, and 117, the first measurement data of the FLOW sensor 116, and the second measurement data of the TEMP sensor 118 that were measured this time, to correct the third measurement data calculated based on the measurement of the pH sensor 113. The correction unit 197 stores the corrected third measurement data in the storage unit 180.

[0210] If no sensors have deviated from the calibration time (No. in step S46), the control unit 190 terminates the process.

[0211] In the flowchart shown in Figure 18, the interpolation unit 198 determines in step S44 whether or not a sensor with a malfunction is included. In this determination, the interpolation unit 198 may also determine whether or not there is only one sensor with a malfunction. If there is only one sensor with a malfunction, the interpolation unit 198 interpolates the measurement data from the malfunctioning sensor using measurement data from multiple sensors acquired in the past and measurement data from other sensors measured this time. If two or more sensors have malfunctions, the interpolation unit 198 stops the measurement by the sensor device 10.

[0212] Furthermore, in the flowchart shown in Figure 18, the correction unit 197 determines in step S46 whether or not a sensor that has deviated from the calibration time is included. In this determination, the correction unit 197 may also determine whether or not there is only one sensor that has deviated from the calibration time. If there is only one sensor that has deviated from the calibration time, the correction unit 197 corrects the measurement data from the sensor that has deviated from the calibration time using measurement data from multiple sensors acquired in the past and measurement data from other sensors measured this time. If two or more sensors have deviated from the calibration time, the correction unit 197 stops the measurement by the sensor device 10.

[0213] Furthermore, in the flowchart shown in Figure 18, the interpolation unit 198 interpolates the measurement data from the faulty sensor in step S45 using measurement data from multiple sensors acquired in the past and measurement data from other sensors measured in the current step. In step S45, the interpolation unit 198 may also interpolate the measurement data from the faulty sensor using measurement data acquired in the past by multiple sensor devices 10 installed in facilities sharing the same water source and measurement data measured in the current step.

[0214] Furthermore, in the flowchart shown in Figure 18, the correction unit 197 corrects the measurement data from the sensor that has deviated from the time of calibration in step S47 using measurement data from multiple sensors acquired in the past and measurement data from other sensors measured in the current measurement. In step S47, the correction unit 197 may also correct the measurement data from the sensor that has deviated from the time of calibration using measurement data acquired in the past by multiple sensor devices 10 installed in facilities sharing the same water source and measurement data measured in the current measurement.

[0215] The first transmitting / receiving unit 192 transmits the first measurement data, the second measurement data, and the third measurement data to the server 20 via the communication unit 120 at predetermined timings. The first transmitting / receiving unit 192 also transmits the first measurement data, the second measurement data, and the third measurement data to other sensor devices 10 installed in the same water treatment facility via the communication unit 120 at predetermined timings. The predetermined timings are, for example, as follows: • Predetermined cycle When information is obtained When information is stored

[0216] <3.2 Operation of Server 20) Server 20 stores data measured by sensor device 10 and provides services that utilize the stored data.

[0217] (Data utilization on server 20: Analysis processing 1) Figure 19 is a flowchart illustrating an example of the operation of server 20 shown in Figure 8.

[0218] The control unit 203 receives from a predetermined user the designation of a water treatment facility associated with that user, and instructions for analysis of the designated water treatment facility (step S51). At this time, the user may select the type of analysis. The control unit 203 may also obtain information about the water treatment facility along with the designation of the water treatment facility from the predetermined user.

[0219] The control unit 203 reads information about the designated water treatment facility from the storage unit 202 using the analysis module 2034. Based on the read information, the analysis module 2034 determines whether the designated water treatment facility has entered into a predetermined contract (step S52). The predetermined contract refers to, for example, a paid contract or a contract equivalent to a paid contract. In the example according to this embodiment, the predetermined contract refers to, for example, a paid contract or a premium contract.

[0220] If the water treatment facility has a specified contract, the analysis module 2034 analyzes the water treatment at the water treatment facility, for example, using the latest version of the first trained model 2027 (step S53). Specifically, the analysis module 2034 inputs information about the specified water treatment facility into the latest version of the first trained model 2027 and outputs an evaluation of the water treatment from the first trained model 2027.

[0221] Analysis module 2034 extracts water treatment facilities similar to a specified water treatment facility based on information read from the water treatment facility. For example, analysis module 2034 extracts water treatment facilities as similar to the specified water treatment facility if they have similar raw water quality, the same type of water treatment facility, similar equipment, perform similar treatments, and use similar chemicals. Analysis module 2034 calculates the degree of similarity to the specified water treatment facility based on, for example, raw water quality, type of water treatment facility, equipment, treatments performed, and chemicals used.

[0222] Analysis module 2034 may also extract similar water treatment facilities by considering the environment of the water treatment facilities. For example, analysis module 2034 may refer to changes in weather, temperature, humidity, wind speed, atmospheric pressure, dust, etc., to extract water treatment facilities located in similar environments. Analysis module 2034 may also calculate the similarity to a specified water treatment facility based on changes in the surrounding environment in which the water treatment facility is located.

[0223] Analysis module 2034 compares a specified water treatment facility with a sample of water treatment facilities.

[0224] The control unit 203 presents the analysis results created in step S53 to the user via the presentation module 2039 (step S54). Specifically, the presentation module 2039 displays the evaluation obtained by the analysis module 2034 and the comparison results with similar water treatment facilities on the terminal device 30 operated by the user.

[0225] Figure 20 is a schematic diagram showing an example of the display on the terminal device 30 used by the user.

[0226] The display example shown in Figure 20 includes a first display area 31 that displays information about a specified water treatment facility and a second display area 32 that displays information about similar water treatment facilities. The first display area 31 includes an evaluation 311 of the water treatment of the water treatment facility, display objects 312 and 313, and an instruction object 314. Display object 312 displays detailed information about the water treatment facility. Display object 313 displays information representing the trend of measurement data. The information representing the trend of measurement data can be displayed in any manner. In this embodiment, for example, it is displayed as a graph. The instruction object 314 is an object for requesting improvement suggestions for the specified water treatment facility.

[0227] The second display area 32 includes an evaluation of the water treatment facility's water treatment process 321, a similarity score 322 with a specified water treatment facility, a display object 323, and an instruction object 324. The display object 323 displays detailed information about the water treatment facility. The instruction object 324 is an object for displaying even more detailed information about the water treatment facility.

[0228] The display module 2039 displays water treatment facilities in the second display area 32 in any order. For example, the display module 2039 displays water treatment facilities in order of plant ID, in order of highest rating, in order of highest similarity, etc.

[0229] When a user requests details about a similar water treatment facility, the presentation module 2039 displays the trend of measurement data for that facility.

[0230] Figure 21 is a schematic diagram showing an example of the display of a terminal device 30 used by a user. In the example shown in Figure 21, a display object 325 that displays the progress of measurement data is displayed in the second display area 32.

[0231] When the control unit 203 receives a request for improvement suggestions from the user, the suggestion module 2037 generates suggestions to improve the evaluation of water treatment (step S55). Specifically, for example, the suggestion module 2037 inputs information about the water treatment facility into a trained model and outputs suggestions for improvement to improve the evaluation of water treatment.

[0232] The control unit 203 presents the user with suggestions to improve the evaluation of water treatment using the presentation module 2039 (step S56). Specifically, the presentation module 2039 displays the suggestions to improve the evaluation of water treatment on the terminal device 30 operated by the user.

[0233] Figure 22 is a schematic diagram showing an example of the display on the terminal device 30 used by the user.

[0234] The display example shown in Figure 22 includes a first display area 31 and a third display area 33 representing proposed improvements to the water treatment facility. The third display area 33 includes display objects 331 to 334 and an instruction object 3321. Display object 331 shows the trend of measurement data as an improvement target. Display objects 332 to 334 show suggestions for improvement. Instruction object 3321 is an object for requesting an estimate of the costs involved in adopting the proposal.

[0235] The presentation module 2039 displays suggestions for improvement to display objects 332-334, such as suggestions for new items to be sensed, suggestions for new locations to be sensed, suggestions for new locations to install the sensor device 10, or suggestions for operational control. The presentation module 2039 also displays the improved operational costs along with each suggestion. Specifically, for example, the presentation module 2039 suggests drainage monitoring to display object 334 and displays the reduction in drainage costs achieved through drainage monitoring.

[0236] Proposal module 2037 generates a quotation in response to a user's request for a quote. Presentation module 2039 then presents the generated quotation to the user.

[0237] In step S52, if the water treatment facility has not entered into a prescribed contract, the analysis module 2034 analyzes the water treatment at the water treatment facility using, for example, an open version of the first pre-trained model 2027 (step S57). Specifically, the analysis module 2034 inputs information about the specified water treatment facility into the open version of the first pre-trained model 2027 and causes the first pre-trained model 2027 to output an evaluation of the water treatment. The analysis module 2034 also extracts water treatment facilities similar to the specified water treatment facility based on the information read out about the water treatment facility.

[0238] The control unit 203 presents the analysis results created in step S57 to the user via the presentation module 2039 (step S58). Specifically, the presentation module 2039 displays the evaluation obtained by the analysis module 2034 and the comparison results with similar water treatment facilities on the terminal device 30 operated by the user.

[0239] When the control unit 203 receives a request for improvement suggestions from the user, the suggestion module 2037 creates suggestions to improve the evaluation of water treatment (step S59). Specifically, for example, the suggestion module 2037 inputs information about the water treatment facility into a trained model and outputs improvement suggestions to improve the evaluation of water treatment. The trained model used at this time may be an older version than the trained model used in step S55.

[0240] The control unit 203 presents the user with suggestions to improve the evaluation of water treatment using the presentation module 2039 (step S510). Specifically, the presentation module 2039 displays the suggestions to improve the evaluation of water treatment on the terminal device 30 operated by the user.

[0241] In the examples shown in Figures 20-22, measurement data acquired by a sensor device 10 installed in a predetermined area is displayed. However, the sensor device 10 that acquires the measurement data may be arbitrarily selected. Furthermore, measurement data acquired by sensor devices 10 installed in multiple areas may also be displayed.

[0242] (Data utilization on server 20: Analysis processing 2) Figure 23 is a flowchart illustrating an example of the operation of server 20 shown in Figure 8.

[0243] The control unit 203 receives instructions from a designated user to specify a water treatment facility associated with that user, and to monitor the specified water treatment facility (step S61).

[0244] The control unit 203 estimates the power consumption of the water treatment facility using the analysis module 2034 (step S62). Specifically, for example, the analysis module 2034 inputs the operating status of the treatment equipment installed in the designated water treatment facility, as well as various sensing information, into a function for calculating power consumption, and calculates the power consumption. The analysis module 2034 may estimate the power consumption of the entire water treatment facility, or it may estimate the power consumption of each treatment equipment installed in the water treatment facility.

[0245] The control unit 203 displays the power consumption amount calculated in step S62 to the user via the display module 2039 (step S63).

[0246] The presentation module 2039 determines whether the calculated power consumption exceeds a preset threshold (step S64). If the calculated power consumption exceeds the preset threshold (Yes in step S64), the presentation module 2039 causes the terminal device 30 to display an alert indicating that the power consumption has exceeded the threshold (step S65). Furthermore, if the calculated power consumption exceeds the preset threshold, the analysis module 2034 estimates the expected impact if this power consumption is maintained. The presentation module 2039 then presents the estimated impact to the user.

[0247] The control unit 203 creates, by means of the proposal module 2037, a countermeasure for reducing the power consumption estimated by the analysis module 2034 (step S66).

[0248] The control unit 203 presents, by means of the presentation module 2039, the countermeasure for reducing the estimated power consumption to the user (step S67). Specifically, the presentation module 2039 causes the countermeasure for reducing the estimated power consumption to be displayed on the terminal device 30 operated by the user.

[0249] FIG. 24 is a schematic diagram showing an example of the display on the terminal device 30 used by the user.

[0250] The display example shown in FIG. 24 includes display objects 341 to 344. In the display object 341, the processors constituting the water treatment facility are displayed. In the display object 342, the power consumption of each of the processors constituting the water treatment facility is displayed. An effect distinguishable from others is applied to the display object 342 whose power consumption exceeds the threshold value. In the example shown in FIG. 24, "Power consumption: W3" is in bold. In the display object 343, an alert indicating that the power consumption exceeds the threshold value is displayed. Further, in the display object 343, the impact that will occur if no countermeasure is taken for the alert is displayed. In the display object 344, the countermeasure for reducing the power consumption is displayed.

[0251] (Data utilization in the server 20: Simulation) FIG. 25 is a flowchart showing an example of the operation of the server 20 shown in FIG. 8.

[0252] The control unit 203 receives, as the designation of the water treatment facility, the input of the design data of the water treatment facility assumed by the user and the water quality data of the water treated in this water treatment facility, and the instruction for simulation from a predetermined user (step S71).

[0253] The control unit 203 determines, by means of the simulation module 2035, whether the user who requested the simulation has concluded a predetermined contract (step S72). If the user has concluded a predetermined contract, the simulation module 2035 estimates, for example, the water treatment in the assumed water treatment facility using the latest version of the second trained model 2028 (step S73). Specifically, the simulation module 2035 inputs the design data of the assumed water treatment facility and the water quality data of the water treated in this water treatment facility into the latest version of the second trained model 2028, and causes the second trained model 2028 to output information for estimating the water treatment. The information for estimating the water treatment includes, for example, the following. · Measurement data at a predetermined position of the water treatment facility · Operating status of the processor · Water usage · Discharge amount of pollutants · Water reuse amount · Operating cost

[0254] The control unit 203 presents the estimation result created in step S73 to the user by means of the presentation module 2039 (step S74). Specifically, the presentation module 2039 causes the estimation result acquired by the simulation module 2035 to be displayed on the terminal device 30 operated by the user.

[0255] FIG. 26 is a schematic diagram showing an example of the display of the terminal device 30 used by the user.

[0256] The display example shown in Figure 26 includes display objects 351-353 and instruction objects 354 and 355. Display object 351 displays the design data of the assumed plant. Display object 352 displays the processors that make up the water treatment facility. Display object 353 displays the power consumption of each processor that makes up the water treatment facility. Instruction object 354 is an object for requesting measurement data acquired by the sensor device 10 installed in the assumed water treatment facility. Instruction object 355 is an object for requesting improvement suggestions for the assumed water treatment facility.

[0257] When the control unit 203 receives a request for improvement suggestions from the user, the suggestion module 2037 creates suggestions to improve the estimation results of water treatment (step S75). Specifically, for example, the suggestion module 2037 creates suggestions for new items to be sensed, new locations to be sensed, new locations to which the sensor device 10 should be newly installed, or suggestions for operational control.

[0258] The control unit 203 presents the user with suggestions to improve the estimated water treatment results using the presentation module 2039 (step S76). Specifically, the presentation module 2039 displays the suggestions to improve the estimated water treatment results on the terminal device 30 operated by the user. The presentation module 2039 also displays the improved operating costs along with each suggestion.

[0259] In step S72, if the user has not entered into a prescribed contract, the simulation module 2035 estimates the water treatment at the assumed water treatment facility using, for example, an open version of the second pre-trained model 2028 (step S77). Specifically, the simulation module 2035 inputs the design data of the assumed water treatment facility and the water quality data of the water treated at this facility into the open version of the second pre-trained model 2028, and outputs information to estimate the water treatment from the second pre-trained model 2028.

[0260] The control unit 203 presents the estimation results created in step S77 to the user via the presentation module 2039 (step S78). Specifically, the presentation module 2039 displays the estimation results obtained by the simulation module 2035 on the terminal device 30 operated by the user.

[0261] When the control unit 203 receives a request for improvement suggestions from the user, the suggestion module 2037 generates suggestions to improve the water treatment estimation results (step S79).

[0262] The control unit 203 presents the user with suggestions to improve the estimated water treatment results using the presentation module 2039 (step S710). Specifically, the presentation module 2039 displays the suggestions to improve the estimated water treatment results on the terminal device 30 operated by the user. The presentation module 2039 also displays the improved operating costs along with each suggestion. (Data utilization on server 20: calibration processing) Figure 27 is a flowchart illustrating an example of the operation of server 20 shown in Figure 8.

[0263] The control unit 203 determines, using the calibration setting module 2036, whether or not there is a sensor device 10 that has a sensor requiring calibration (step S81). Specifically, the calibration setting module 2036 determines, for example, whether or not a sensor is in a state requiring calibration based on information obtained from measurements at the sensor device 10. In other words, the calibration setting module 2036 determines that a sensor requires calibration if, for example, there is a sensor whose measured value changes in a different trend than other sensors over time. The calibration setting module 2036 may also determine that a sensor is in a state requiring calibration if a predetermined period of time has elapsed since the last calibration. The calibration setting module 2036 may also determine that a sensor provided in the sensor device 10 is in a state requiring calibration if a calibration request is received from the sensor device 10.

[0264] If there is a sensor device 10 that has a sensor that requires calibration (Yes in step S81), the calibration setting module 2036 calculates calibration information for that sensor (step S82). Specifically, the calibration setting module 2036 calculates calibration information from, for example, the first measurement data, second measurement data, and third measurement data acquired by the sensor device 10, which are stored in the measurement table 2024, and the calibration information set in the past, which is stored in the calibration table 2025. For example, the calibration setting module 2036 takes the measurement information from the sensor device 10 measured in the past as input data and the calibration information set in the past as ground truth output data, and inputs the latest first measurement data, second measurement data, and third measurement data into a trained model to obtain calibration information.

[0265] The control unit 203 transmits the calculated calibration information to the sensor device 10 having the sensor that is determined to require calibration, via the transmission control module 2032 (step S83).

[0266] As described above, in the above embodiment, the control unit 190 acquires measurement data from multiple sensors 111 to 118, each measuring different items, using the second transmitting / receiving unit 193. The control unit 190 uses the estimation unit 196 to estimate whether or not there is an abnormality in any of the multiple sensors, based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items. The control unit 190 uses the correction unit 197 or the interpolation unit 198 to correct the measurement data measured by the sensor that is estimated to have an abnormality, based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items. This makes it possible to estimate in real time which sensor among the sensors 111 to 118 has an abnormality. It also makes it possible to correct the measurement data from the sensor that has an abnormality in real time. Therefore, when an abnormality occurs in a sensor of the sensor device 10, it is possible to suppress the impact of the abnormality on the measurement data. Furthermore, even if an abnormality occurs in a sensor, it is no longer necessary to immediately stop the measurement.

[0267] Therefore, in the sensor device 10 having multiple sensors that measure different items according to this embodiment, the measurement items can be measured accurately.

[0268] Furthermore, in the above embodiment, the estimation unit 196 estimates whether any of the multiple sensors is malfunctioning based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items. The interpolation unit 198 interpolates the measurement data measured by the sensor estimated to be malfunctioning based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items. This makes it possible to estimate in real time which sensor among sensors 111 to 118 has malfunctioned. It also makes it possible to interpolate the measurement data from the malfunctioning sensor in real time. Therefore, when a sensor in the sensor device 10 malfunctions, the impact of the malfunction on the measurement data can be minimized. In addition, even if a sensor malfunctions, it is no longer necessary to immediately stop the measurement.

[0269] Furthermore, in the above embodiment, the estimation unit 196 estimates whether any of the multiple sensors have deviated from the calibration time based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items. The correction unit 197 corrects the measurement data measured by the sensor that is estimated to have deviated from the calibration time based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items. This makes it possible to estimate in real time which of the sensors 111 to 118 has deviated from the calibration time. It also makes it possible to correct the measurement data from the sensor that has deviated from the calibration time in real time. Therefore, when a deviation occurs in the sensors of the sensor device 10 from the calibration time, it is possible to minimize the impact on the measurement data.

[0270] Further, in the above embodiment, the control unit 190 acquires, by the first transmission / reception unit 192, measurement data of a plurality of items measured by another sensor device 10 that measures water from the same water source. The estimation unit 196 estimates whether there is an abnormality in any of the plurality of sensors based on the measurement data of the plurality of items acquired, the measurement data of the plurality of items acquired from another sensor device 10, the measurement data of the plurality of items acquired in the past, and the measurement data of the plurality of items acquired from another sensor device 10 in the past. Thereby, in order to compare the sensors of the sensor device 10 that measures water from the same water source, which are expected to exhibit the same behavior, it becomes possible to highly accurately detect an abnormality occurring in the sensor.

[0271] Further, in the above embodiment, the correction unit 197 or the complementation unit 198 corrects the measurement data measured by the sensor estimated to be abnormal based on the measurement data of the plurality of items acquired, the measurement data of the plurality of items acquired from another sensor device 10, the measurement data of the plurality of items acquired in the past, and the measurement data of the plurality of items acquired from another sensor device 10 in the past. Thereby, in order to compare the sensors of the sensor device 10 that measures water from the same water source, which are expected to exhibit the same behavior, it becomes possible to highly accurately correct the influence of an abnormality occurring in the sensor.

[0272] Further, in the above embodiment, the estimation unit 196 estimates the presence or absence of an abnormality by inputting the measurement data of the plurality of items acquired into a learned model that has learned the measurement data of the plurality of items acquired in the past as input data and the determination of the occurrence of an abnormality based on the measurement data as correct answer output data. Thereby, it becomes possible to further improve the accuracy of estimating the occurrence of an abnormality.

[0273] <4 Modification Example> In the above embodiment, the sensor device 10 is described as having a touch panel 1119. However, the sensor device 10 does not have to have a touch panel 1119. The control unit 190 may transmit information related to processing in the control unit 190 to a terminal device 30 operated by the user via the first transmitting / receiving unit 192. The terminal device 30 displays the information transmitted from the sensor device 10 on the display of the terminal device 30. The sensor device 10 may also receive user operations via the terminal device 30 via the first transmitting / receiving unit 192.

[0274] The terminal device 30 installs, for example, an application for coordinating with the sensor device 10. This enables the terminal device 30 to display information related to processing by the control unit 190 on its display, and to accept user requests for operation to the sensor device 10.

[0275] The pre-trained models described in the above embodiments may be used in different versions depending on the contract, with or without explanation. Furthermore, multiple types of the first pre-trained model 2027 may be created depending on the criteria for analysis. The types of the first pre-trained model 2027 used may also differ depending on the contract.

[0276] In the above embodiment, the sensor device 10 was described as having sensors capable of measuring different items. However, any of the sensors in the sensor device 10 may be configured to measure the same item. For example, if the value of one sensor changes and becomes ineligible for the value of the other sensor, the sensor device 10 will determine that an abnormality has occurred in one of the sensors. This makes it possible to determine more accurately whether or not an abnormality has occurred in a sensor.

[0277] In the above embodiment, if the estimation unit 196 estimates that the sensors that detected the measured information include sensors that have deviated from their calibration values, the sensor device 10 may request calibration information for those sensors from the server 20. At this time, the deviation from calibration values ​​may differ depending on whether the measurement value is being corrected or calibration information is being requested. For example, the deviation may be set to be larger when calibration information is being requested than when the measurement value is being corrected. In this way, the estimation unit 196 estimates which sensors among the multiple sensors require calibration based on measurement data of multiple items acquired from the multiple sensors and measurement data of multiple items acquired in the past. The first transmitting / receiving unit 192 then requests calibration information for the estimated sensors that require calibration from the server 20. As a result, the sensor device 10 automatically determines whether calibration is necessary and requests calibration information from the server.

[0278] <5 Basic Hardware Configuration of a Computer> Figure 28 is a block diagram showing the basic hardware configuration of computer 90. Computer 90 includes at least a processor 91, main memory 92, auxiliary memory 93, and a communication interface 99. These are electrically connected to each other by a bus.

[0279] The processor 91 is hardware for executing the instruction set written in the program. The processor 91 consists of an arithmetic unit, registers, peripheral circuits, etc.

[0280] Main memory 92 is used to temporarily store programs and data processed by programs, etc. For example, it is a volatile memory such as DRAM (Dynamic Random Access Memory).

[0281] Auxiliary storage device 93 is a storage device for storing data and programs. Examples include flash memory, HDD (Hard Disc Drive), magneto-optical disk, CD-ROM, DVD-ROM, semiconductor memory, etc.

[0282] A communication IF99 is an interface for inputting and outputting signals for communication with other computers via a network using wired or wireless communication standards. A network consists of various mobile communication systems, such as the internet, LANs, and wireless base stations. For example, a network includes 3G, 4G, and 5G mobile communication systems, LTE (Long Term Evolution), and wireless networks that can connect to the internet via designated access points (e.g., Wi-Fi®). When connecting wirelessly, communication protocols include, for example, Z-Wave®, ZigBee®, and Bluetooth®. When connecting via a wired connection, the network also includes connections made directly via USB (Universal Serial Bus) cables, etc.

[0283] Furthermore, by distributing all or part of each hardware configuration across multiple computers 90 and connecting them to each other via a network, a computer 90 can be virtually realized. Thus, the concept of computer 90 includes not only a computer 90 housed in a single enclosure or case, but also a virtualized computer system.

[0284] <Basic Functional Configuration of Computer 90> The functional configuration of the computer realized by the basic hardware configuration of computer 90 shown in Figure 28 will be explained. The computer comprises at least one functional unit: a control unit, a memory unit, and a communication unit.

[0285] Furthermore, the functional units of computer 90 can also be realized by distributing all or part of each functional unit across multiple computers 90 interconnected via a network. The concept of computer 90 includes not only a single computer 90 but also a virtualized computer system.

[0286] The control unit is realized when the processor 91 reads various programs stored in the auxiliary storage device 93, loads them into the main memory device 92, and executes processing according to those programs. The control unit can realize various functional units that perform information processing depending on the type of program. In this way, the computer is realized as an information processing device that performs information processing.

[0287] The memory unit is implemented by a main memory 92 and an auxiliary memory 93. The memory unit stores data, various programs, and various databases. The processor 91 can also reserve memory areas corresponding to the memory unit in the main memory 92 or the auxiliary memory 93 according to the program. The control unit can also cause the processor 91 to perform operations such as adding, updating, and deleting data stored in the memory unit according to the various programs.

[0288] A database, specifically a relational database, is used to manage and link data sets called tables, which are structurally defined by rows and columns. In a database, tables are called tables, the columns of a table are called columns, and the rows of a table are called records. In a relational database, relationships can be established and linked between tables. Typically, each table has a key column to uniquely identify records, but setting a key on a column is not mandatory. The control unit can instruct the processor 91 to add, delete, or update records in specific tables stored in the memory unit according to various programs.

[0289] The communication unit is implemented by the communication IF99. The communication unit implements the function of communicating with other computers 90 via the network. The communication unit can receive information transmitted from other computers 90 and input it to the control unit. The control unit can cause the processor 91 to perform information processing on the received information according to various programs. The communication unit can also transmit information output from the control unit to other computers 90.

[0290] While several embodiments of this disclosure have been described above, these embodiments can be implemented in a variety of other forms, and various omissions, substitutions, and modifications are permitted without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents.

[0291] <Note> The details described in each of the above embodiments are noted below.

[0292] (Note 1) A program for execution on a computer having a processor and memory, the program causing the processor to perform the following steps: acquiring measurement data from a plurality of sensors, each measuring a different item; estimating whether or not there is an abnormality in any of the plurality of sensors based on the acquired measurement data of the plurality of items and previously acquired measurement data of the plurality of items; and correcting the measurement data measured by the sensor that is estimated to be abnormal based on the acquired measurement data of the plurality of items and previously acquired measurement data of the plurality of items. (Note 2) The program described in (Appendix 1) includes the following steps: in the estimation step, it estimates whether any of the multiple sensors are malfunctioning based on the acquired measurement data for multiple items and the measurement data for multiple items acquired in the past; and in the correction step, it supplements the measurement data measured by the sensor that is estimated to be malfunctioning based on the acquired measurement data for multiple items and the measurement data for multiple items acquired in the past. (Note 3) The program described in (Appendix 1) or (Appendix 2), wherein in the estimation step, it estimates whether any of the multiple sensors have deviated from the calibration date based on the acquired measurement data of multiple items and the measurement data of multiple items acquired in the past, and in the correction step, it corrects the measurement data measured by the sensor that is estimated to have deviated from the calibration date based on the acquired measurement data of multiple items and the measurement data of multiple items acquired in the past. (Note 4) A program as described in any of (Appendix 1) to (Appendix 3), wherein the processor performs the step of acquiring measurement data for multiple items measured by other sensor devices that measure water from the same water source, and in the step of estimation, estimates whether or not there is a malfunction in any of the multiple sensors based on the acquired measurement data for multiple items, measurement data for multiple items acquired from other sensor devices, measurement data for multiple items acquired in the past, and measurement data for multiple items acquired in the past from other sensor devices. (Note 5) The program described in (Appendix 4) corrects the measurement data measured by the sensor suspected to have an abnormality in the correction step, based on the acquired measurement data for multiple items, the acquired measurement data for multiple items from other sensor devices, the previously acquired measurement data for multiple items, and the previously acquired measurement data for multiple items from other sensor devices. (Note 6) In the estimation step, the program estimates whether or not an anomaly has occurred by inputting the acquired measurement data for multiple items into a trained model that has been trained using previously acquired measurement data for multiple items as input data and judgments of anomaly occurrence based on said measurement data as the correct output data. (A program as described in any of Appendix 1 to Appendix 5.) (Note 7) A method to be performed on a computer having a processor and memory, wherein the processor performs the steps of: acquiring measurement data from a plurality of sensors, each measuring a different item; estimating whether any of the plurality of sensors is abnormal based on the acquired measurement data of the plurality of items and previously acquired measurement data of the plurality of items; and correcting the measurement data measured by the sensor that is estimated to be abnormal based on the acquired measurement data of the plurality of items and previously acquired measurement data of the plurality of items. (Note 8) An information processing device comprising a control unit and a storage unit, wherein the control unit performs the steps of: acquiring measurement data from a plurality of sensors, each measuring a different item; estimating whether or not there is an abnormality in any of the plurality of sensors based on the acquired measurement data for the plurality of items and measurement data for the plurality of items acquired in the past; and correcting the measurement data measured by the sensor estimated to be abnormal based on the acquired measurement data for the plurality of items and measurement data for the plurality of items acquired in the past. (Note 9) A system that performs the following actions: acquiring measurement data from multiple sensors, each measuring a different item; estimating whether any of the multiple sensors are abnormal based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items; and correcting the measurement data measured by the sensor estimated to be abnormal based on the acquired measurement data for multiple items and previously acquired measurement data for multiple items. [Explanation of Symbols]

[0293] 1... System 10...Sensor device 11…Cabinet 11a... Pipe connection hole 11b...Pipe connection hole 11c... Pipe connection hole 11d... Air intake 11e... Exhaust port 111...EC Sensor 112...FCL sensor 113... pH sensor 1131...pH electrode 1132... Circuit board 11321…CPU 11322...memory 113221... Pre-trained model 11323… Amplifier 11324… Converter 11325... Input / Output Interface 11326... Communications Department 114...ORP sensor 115...NO3 sensor 115a...gripping part 115b…Tsubabe 115c...Measuring part 116…FLOW sensor 117…TUR Sensor 1171... Shell 118...TEMP sensor 119... Valve 1110... Valve 1111...Flow Cell 11111~11113... Measurement port 11112a...Cylinder part 11112b…Cylinder part 11113b…Cylinder part 11113c…hole 11114… Water supply channel 11115... Air venting channel 1112... Control box 1113…Terminal block 1114…Terminal block 1115... Air vent valve 11151...Socket 1116... Drain valve 1117... Flow control valve 1118... Air filter 1119... Touch panel 11191…Touch-sensitive device 11192…Display 1120~1122... Fittings 12…Lid 120... Communications Department 131…Touch-sensitive devices 141…Display 180...Storage section 181…Calibration information 182...Measurement Information 183...Sensor measurement information 184...Measurement Information 185... Pre-trained model 190... Control Unit 191... Operation reception desk 192...First Transmitter / Receiver Unit 193...Second Transceiver Unit 194…Proofreading Department 195...Calculation section 196...Estimation part 197... Correction section 198... Supplementary section 199…Presentation Control Unit 20... Server 201... Communications Department 202...Storage section 2021...Plant Table 2022…Installation Table 2023... Plant Environment Table 2024... Measurement Table 2025…Correction Table 2026…Model Table 2027…First pre-trained model 2028…Second pre-trained model 203... Control Unit 2031…Receiver control module 2032…Transmission control module 2033…Memory control module 2034…Analysis Module 2035… Simulation Module 2036…Calibration settings module 2037… Proposed Module 2038…Learning Module 2039…Presented Module 30…Terminal device 80…Network 90... Computer 91… Processor 92…Storage device 93…Auxiliary storage device 99...Communication IF

Claims

1. A program to be executed by a computer having a processor and memory, wherein the program is to be executed by the processor, The process involves acquiring measurement data from multiple sensors, each measuring a different item, and A step of estimating whether or not there is an abnormality in any of the multiple sensors, based on the acquired measurement data of multiple items and the measurement data of multiple items acquired in the past, The step of correcting measurement data obtained from a sensor suspected of having an abnormality, based on the acquired measurement data for multiple items and measurement data for multiple items acquired in the past. A program that executes the command.

2. In the estimation step described above, whether or not any of the multiple sensors is malfunctioning is estimated based on the acquired measurement data of multiple items and the measurement data of multiple items acquired in the past. The program according to claim 1, wherein in the correction step, measurement data measured by a sensor that is presumed to be malfunctioning is supplemented based on acquired measurement data of multiple items and measurement data of multiple items acquired in the past.

3. In the estimation step described above, whether or not a deviation from the time of calibration has occurred in any of the multiple sensors is estimated based on the acquired measurement data of multiple items and the measurement data of multiple items acquired in the past. The program according to claim 1 or 2, wherein in the correction step, measurement data measured by a sensor that is estimated to have deviated from the time of calibration is corrected based on acquired measurement data of multiple items and measurement data of multiple items acquired in the past.

4. The processor is instructed to perform the step of acquiring measurement data for multiple items measured by other sensor devices that measure water from the same water source. The program according to any one of claims 1 to 3, wherein in the estimation step, it estimates whether or not there is an abnormality in any of the multiple sensors based on the acquired measurement data of multiple items, the measurement data of multiple items acquired from the other sensor device, the measurement data of multiple items acquired in the past, and the measurement data of multiple items acquired in the past from the other sensor device.

5. The program according to claim 4, wherein in the correction step, the program corrects the measurement data measured by the sensor that is presumed to have an abnormality based on the acquired measurement data of multiple items, the measurement data of multiple items acquired from the other sensor device, the measurement data of multiple items acquired in the past, and the measurement data of multiple items acquired in the past from the other sensor device.

6. The program according to any one of claims 1 to 5, wherein in the estimation step, the program estimates whether or not an anomaly has occurred by inputting the acquired measurement data of multiple items into a trained model that has been trained using previously acquired measurement data of multiple items as input data and judgments of anomaly occurrence based on said measurement data as correct output data.

7. A method to be performed on a computer comprising a processor and memory, wherein the processor The process involves acquiring measurement data from multiple sensors, each measuring a different item, and A step of estimating whether or not there is an abnormality in any of the multiple sensors, based on the acquired measurement data of multiple items and the measurement data of multiple items acquired in the past, The step of correcting measurement data obtained from a sensor suspected of having an abnormality, based on the acquired measurement data for multiple items and measurement data for multiple items acquired in the past. How to do it.

8. An information processing apparatus comprising a control unit and a storage unit, wherein the control unit is The process involves acquiring measurement data from multiple sensors, each measuring a different item, and A step of estimating whether or not there is an abnormality in any of the multiple sensors, based on the acquired measurement data of multiple items and the measurement data of multiple items acquired in the past, The step of correcting measurement data obtained from a sensor suspected of having an abnormality, based on the acquired measurement data for multiple items and measurement data for multiple items acquired in the past. An information processing device that performs the following actions.

9. A means of acquiring measurement data from multiple sensors, each measuring a different item, A means for estimating whether or not there is an abnormality in any of multiple sensors, based on acquired measurement data of multiple items and previously acquired measurement data of multiple items, A means for correcting measurement data obtained from a sensor suspected of having an abnormality, based on the acquired measurement data for multiple items and measurement data for multiple items acquired in the past. A system that executes this process.

Images