System, information processing device, program, and information processing method
The system addresses the lack of power consumption consideration in conventional maintenance management by predicting power generation and consumption, enabling optimized operation policies for facilities with photovoltaic panels.
Patent Information
- Application Number
- JP2025078376
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-05-09
- Publication Date
- 2026-11-19
AI Technical Summary
Conventional maintenance management systems for photovoltaic power generation panels do not consider power consumption within facilities equipped with these panels, especially when installed on residential roofs, necessitating an optimization of operation policies that account for both external environments and internal power usage.
A system comprising a prediction module for forecasting power generation and consumption, and an optimization module for determining optimal operating policies, utilizing actual data and weather information to enhance maintenance management.
Enables optimized operation policies that consider both power generation and consumption, improving maintenance efficiency and effectiveness in facilities with photovoltaic panels.
Smart Images

Figure 2026182272000001_ABST
Abstract
Description
Technical Field
[0001] Embodiments of the present disclosure relate to a system, an information processing apparatus, a program, and an information processing method.
Background Art
[0002] Maintenance management systems for performing maintenance of photovoltaic power generation panels have been developed. Document 1 discloses a maintenance management apparatus that diagnoses the safety or service life of photovoltaic power generation facilities, accumulates information related to the performance, procurement, or contractors of devices that control the photovoltaic power generation panels, and provides diagnostic data, facility data, or procurement information to a server of a service provider.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] On the other hand, in conventional maintenance management systems, photovoltaic panels and the maintenance management system are provided as a single package, and the power consumption in facilities equipped with photovoltaic panels has not been taken into consideration. In recent years, photovoltaic panels are sometimes installed on the roofs of houses, and in such scenarios, there is a need to optimize the operation policy for each facility while taking into account the external environment of the facility and the power consumption within the facility.
[0005] The problem to be solved by the present disclosure is to provide a system that optimizes the operation policy of facilities including power generation equipment, while taking into account the external environment of the facility and the internal power consumption.
Means for Solving the Problems
[0006] In one aspect of this disclosure, a system is provided comprising: a prediction module that predicts future power generation and power consumption in a facility including power generation equipment; and an optimization module that estimates the optimal operating policy for the equipment in the facility based on the predicted power generation and power consumption, wherein the prediction module predicts the power generation based on actual power generation and power consumption data in the facility and weather data relating to the location where the equipment is installed. [Brief explanation of the drawing]
[0007] [Figure 1] This figure shows the overall configuration of the maintenance management system 1 according to the embodiment. [Figure 2] This diagram shows the functional configuration of the data management server 200. [Figure 3] Figure 3A shows the management screen 410 displayed on the terminal device 400's display. Figure 3B shows the equipment list screen 420 displayed on the terminal device 400's display. [Figure 4] Figure 4A shows the equipment details screen 430 displayed on the terminal device 400's display. Figure 4B shows the equipment details editing screen 440 displayed on the terminal device 400's display. [Figure 5] Figure 5A shows the usage management screen 450 displayed on the terminal device 400's display. Figure 5B shows the usage details screen 460 displayed on the terminal device 400's display. Figure 5C shows the details editing screen 470 displayed on the terminal device 400's display. [Figure 6] Figure 6A shows the management screen 410 displayed on the terminal device 400's display. Figure 6B shows the organization management screen 480 displayed on the terminal device 400's display. [Figure 7] Figure 7A shows the user information details screen 490 displayed on the terminal device 400's display. Figure 7B shows the user information editing screen 495 displayed on the terminal device 400's display. [Figure 8]This is a process flow diagram showing the first half of the data access processing procedure using tag 110. [Figure 9] This is a process flow diagram showing the second half of the data access processing procedure using tag 110. [Figure 10] This is a diagram showing the functional configuration of the data analysis server 300. [Figure 11] This is a diagram showing the data included in the failure detection model 380. [Figure 12] This is a flowchart showing the processing flow of the failure detection module 320. [Figure 13] This is a diagram explaining the abnormality detection learning process. [Figure 14] This is a diagram explaining the waveform classification learning process. [Figure 15] This is a diagram explaining the failure classification learning process. [Figure 16] This is a diagram explaining the failure mode learning process. [Figure 17] This is a diagram explaining the explanatory variable learning process. [Figure 18] This is a diagram explaining the failure detection learning process. [Figure 19] This is a diagram explaining the inference processing by the failure detection module 320. [Figure 20] This is a diagram explaining failure mode inference. [Figure 21] This is a flowchart showing the processing flow of the prediction module 330. [Figure 22] This is a diagram explaining the explanatory variable learning process. [Figure 23] This is a diagram explaining the prediction learning process. [Figure 24] This is a diagram explaining the inference processing by the prediction module 330. [Figure 25] This is a diagram explaining the processing by the optimization module 340. [Figure 26] This is a block diagram showing the hardware configuration of the information processing apparatus. [Mode for Carrying Out the Invention]
[0008] <1. Embodiment> Hereinafter, a maintenance management system 1 as an embodiment of the present disclosure will be described with reference to the drawings. In the specification of the present application and each drawing, elements identical to those already described are designated by the same reference numerals, and detailed descriptions thereof will not be repeated. In the present disclosure, an example of a user of the maintenance management system 1 may be a maintenance management user who uses the maintenance management system 1 to perform maintenance management of facilities such as solar panels. The maintenance management user may perform maintenance management of the facilities of their own company, or may be entrusted with the maintenance management of facilities of another company. Further, another example of the user of the maintenance management system 1 may be a system management user who manages the maintenance management system 1. The maintenance management user and the system management user may be the same or different. The term "user" in the following description may include both the maintenance management user and the system management user.
[0009] (1.1. Overall Configuration of Maintenance Management System 1) The functional configuration of the maintenance management system 1 according to the embodiment will be described with reference to Figure 1. The maintenance management system 1 is a system that performs maintenance management of facilities. An example of the facilities may be a solar panel P. As an example, the solar panel P is installed in a solar power plant F. As another example, the solar panel P is installed on the roof of a house H. The maintenance management system 1 detects a failure occurring in the solar panel P, and estimates the cause of the failure. In addition, the maintenance management system 1 grasps the actual power generation amount of the solar panel P, and predicts the future power generation amount. Further, when the solar panel P is installed in the house H, the maintenance management system 1 manages the power consumption of household electric appliances (hereinafter referred to as home electric appliances) installed in the house H.
[0010] The maintenance management system 1 comprises an IoT device 100, a data management server 200, a data analysis server 300, and a terminal device 400. The IoT device 100, the data management server 200, the data analysis server 300, and the terminal device 400 are connected to a network N such as the Internet and are configured to communicate with each other.
[0011] The IoT device 100 is a device attached to equipment that transmits and receives data via a network N. The IoT device 100 includes a detection unit and a data storage unit. The detection unit detects information regarding the status of the equipment or the status of the environment in which the equipment is installed. For example, the detection unit detects actual data on power generation by solar panels P and power consumption at the facility where the solar panels P are installed. The data storage unit is attached to the equipment and stores data related to the equipment. In this embodiment, the detection unit includes a current sensor 120 and a surveillance camera 130. The data storage unit also includes a tag 110. However, the detection unit and the data storage unit are not limited to these examples. Specifically, the detection unit may be a sensor that detects other physical quantities such as temperature, atmospheric pressure, humidity, water pressure, and voltage.
[0012] Tag 110 is, for example, an RFID (Radio Frequency Identification) tag that uses wireless communication to store and / or transmit data. Tag 110 receives radio waves (induction signals) sent from the terminal device 400 operated by the user and returns the information it holds to the terminal device 400. The information returned includes information for accessing virtual storage stored in the data management server 200.
[0013] The current sensor 120 is a sensor for measuring the current in a predetermined electrical circuit, and detects the magnitude and change of the current flowing through the electrical circuit. The measured current value is transmitted to the data management server 200. The current sensor 120 can be any known type, such as a Hall effect sensor, shunt resistor type, or current transformer, as appropriate.
[0014] The surveillance camera 130 is installed around the facility and acquires image data. From the image data acquired by the surveillance camera 130, it is possible to detect damage caused by wild animals, theft, natural disasters, etc. The image data acquired by the surveillance camera 130 is transmitted to the data management server 200.
[0015] The data management server 200 is an information processing device that performs data management, as an example. The hardware configuration of the information processing device will be described later. The data managed by the data management server 200 includes data related to equipment, specifically data related to the status of equipment detected by the detection unit, and data related to the status of the environment in which the equipment is installed, as detected by the detection unit. Details of the processing performed by the data management server 200 will be described later.
[0016] The data analysis server 300 estimates the nature and cause of failures that occur in the solar panels P, and makes future predictions regarding the operation of the equipment. Specifically, the data analysis server 300 uses a trained model, which has been developed through machine learning based on historical failure data, to estimate the nature and cause of failures that occur in the solar panels P. In addition, the data analysis server 300 uses a trained model, which has been developed through machine learning based on historical power generation data, to predict power generation. Details of the processing performed by the data analysis server 300 will be described later.
[0017] The terminal device 400 is an information processing device equipped with input means and output means, which can be operated by users and administrators. The terminal device 400 may be a general-purpose or dedicated personal computer, a smart device, etc. Smart devices include tablet terminals, smartphones, smart glasses, smartwatches, etc. If the terminal device 400 is a tablet terminal or a smartphone, the terminal device 400 has a display that functions as input means and output means. The terminal device 400 accepts data input by the user's touch operation on the display.
[0018] The maintenance management system 1 may also acquire data from an artificial satellite S. This data from the artificial satellite S may include, for example, weather data, or satellite imagery of solar panels P. This data may also be stored in the data management server 200.
[0019] (1.2. Data Management Server 200) (1.2.1. Functional configuration of data management server 200) Referring to Figure 2, the functional configuration of the data management server 200 will be explained. The data management server 200 comprises a control unit 210 and a storage unit 220. The control unit 210 acts as a processor and controls the processing of the data management server 200.
[0020] The storage unit 220 stores programs and data necessary for processing by the control unit 210, as well as data obtained through processing by the control unit 210. For example, the storage unit 220 stores virtual storage 230.
[0021] The virtual storage 230 stores the IoT data management table 230a. The IoT data management table 230a includes, as an example, the fields "Management Number", "T1 Management Number", "T1 Address", "T2 Management Number", "T2 Address", "T3 Management Number", and "T3 Address".
[0022] The field "Management Number" is data used to uniquely identify a record in IoT data management table 230a and is the primary key field of IoT data management table 230a. The field "T1 Management Number" is a foreign key field used to reference the primary key field "Management Number" in IoT data table T1 stored in data center S1. The field "T1 Address" is address information used to access IoT data table T1 stored in data center S1.
[0023] The field "T2 Management Number" is a foreign key field used to reference the primary key field "Management Number" in the IoT data table T2 stored in data center S2. The field "T2 Address" is address information used to access the IoT data table T2 stored in data center S2.
[0024] The item "T3 Management Number" is a foreign key item used to reference the primary key item "Management Number" in IoT data table T3 stored in the information processing device within the factory PL. The item "T3 Address" is address information used to access IoT data table T3 stored in the information processing device within the factory PL.
[0025] In this way, the maintenance management system 1 centrally manages the information detected by the current sensor 120 and the surveillance camera 130 as detection units, as well as the information held by the tag 110 as a data holding unit, in virtual storage. That is, the actual data is stored in the information processing device in data center S1, data center S2, or within the factory PL, and the data management server 200 stores the address information and key information for accessing that external data, thereby configuring the virtual storage 230. With this specification, even if the information detected by the current sensor 120 and the surveillance camera 130 as detection units, as well as the information held by the tag 110 as a data holding unit, is stored in various data sensors and information processing devices, it is centrally managed in the virtual storage 230 stored in the data management server 200, so users can easily obtain the data necessary for the maintenance management of existing equipment.
[0026] (1.2.2. Example screens in data management) The following describes examples of screens displayed on the terminal device 400 for data management, with reference to Figures 3 to 7. Figure 3A shows the management screen 410 displayed on the terminal device 400's display. As shown in Figure 3A, the management screen 410 has a menu list 411 and an organization page 412. The menu list 411 displays a list of screens that can be accessed from the management screen 410. The organization page 412 is the home screen of the management screen 410. The organization page 412 displays user information 413 registered in the organization. When a user selects the equipment list icon 411a in the menu list 411, the equipment list screen 420 is displayed.
[0027] Figure 3B shows the equipment list screen 420 displayed on the terminal device 400. As shown in Figure 3B, the equipment list screen 420 has an equipment list 421, an equipment details button 422, and an application editing button 423. The equipment list 421 displays a list of equipment to which tags 110 are attached. The equipment details button 422 is a button for displaying detailed information about the equipment. When the user presses the equipment details button 422, the equipment details screen 430 is displayed. The application editing button 423 is a button for editing the application of tag 110 (what information is associated with tag 110). When the user presses the application editing button 423, the application management screen 450 is displayed.
[0028] Figure 4A shows the equipment details screen 430 displayed on the terminal device 400. As shown in Figure 4A, the equipment details screen 430 has equipment details information 431 and an equipment details edit button 432. The equipment details information 431 is information about the details of the equipment. The equipment details edit button 432 is a button for editing the details of the equipment. When the user presses the equipment details edit button 432, the equipment details editing screen 440 is displayed.
[0029] Figure 4B shows the equipment details editing screen 440 displayed on the terminal device 400's display. As shown in Figure 4B, the equipment details editing screen 440 has a display settings list 441. The display settings list 441 is a list of settings for whether or not to display the information of the equipment on the management screen. The display settings list 441 includes a public setting checkbox 442. When a user checks the public setting checkbox 442, the information of the equipment is made public on the network N, and authorized users can access and view it from outside. On the other hand, when a user unchecks the public setting checkbox 442, users cannot access and view it from outside. In this way, the maintenance management system 1 can set whether or not to make the information of each piece of equipment to which a tag 110 is attached public on the network N.
[0030] Figure 5A shows the usage management screen 450 displayed on the terminal device 400. As shown in Figure 5A, the usage management screen 450 has a usage list 451. The usage list 451 is a list of uses for the tag 110. In the maintenance management system 1, multiple uses can be set for what kind of information to associate with the tag 110. The usage list 451 includes a details button 452. When the user presses the details button 452, the usage details screen 460 is displayed.
[0031] Figure 5B shows the application details screen 460 displayed on the terminal device 400. As shown in Figure 5B, the application details screen 460 has an application item list 461. The application item list 461 is a list of data items set for each application. The application item list 461 has an edit button 462. When the user presses the edit button 462, the detailed editing screen 470 is displayed.
[0032] Figure 5C shows the detailed editing screen 470 displayed on the terminal device 400. As shown in Figure 5C, the detailed editing screen 470 has item details 471. Item details 471 is information about the details of data items set for each purpose. Item details 471 includes a guest hide checkbox 472. When a user checks the guest hide checkbox 472, the item is hidden from guest users with limited access rights (for example, a construction company employee temporarily using the maintenance management system 1). On the other hand, when a user unchecks the guest hide checkbox 472, the item is displayed to guest users. In this way, the maintenance management system 1 allows setting whether or not to display each data item associated with the tag 110 to guest users.
[0033] Figure 6A shows the management screen 410 displayed on the terminal device 400. As shown in Figure 6A, the management screen 410 has an organization information pull-down list 414. The organization information pull-down list 414 is a list selected when managing, switching, adding, etc., organization information. The organization information pull-down list 414 includes an organization management icon 414a. When the user presses the organization management icon 414a, the organization management screen 480 is displayed.
[0034] Figure 6B shows the organization management screen 480 displayed on the terminal device 400. As shown in Figure 6B, the organization management screen 480 has a user list 481. The user list 481 is a list of users belonging to the organization. The user list 481 has a details button 482. When a user presses the details button 482, the user information details screen 490 is displayed.
[0035] Figure 7A shows the user information details screen 490 displayed on the terminal device 400. As shown in Figure 7A, the user information details screen 490 has a user information details list 491 and a user information edit button 492. The user information details list 491 is a list of detailed information about the user. The user information edit button 492 is a button for editing user information. When the user presses the user information edit button 492, the user information edit screen 495 is displayed.
[0036] Figure 7B shows the user information editing screen 495 displayed on the terminal device 400. As shown in Figure 7B, the user information details list 491 on the user information editing screen 495 includes an inspection registration permission checkbox 496. When a user checks the inspection registration permission checkbox 496, they can register and update information after inspecting the equipment, regardless of their access privileges (i.e., even if they are a guest user as shown in Figure 7B). On the other hand, when a user unchecks the inspection registration permission checkbox 496, that user cannot register or update information after inspecting the equipment. In this way, the maintenance management system 1 can grant each user permissions regarding the registration and updating of equipment-related information.
[0037] (1.2.3. Data access using tag 110) Refer to Figures 8 and 9 to explain the processing flow for data access using tag 110. As shown in Figure 8, in step S210, the terminal device 400 transmits a guidance signal to the tag 110. In step S110, the tag 110 receives the guidance signal transmitted from the terminal device 400.
[0038] In step S115, the tag 110 transmits access information (for example, the address of the virtual storage 230 stored in the data management server 200) to the terminal device 400 based on the received guidance signal. In step S215, the terminal device 400 receives the access information transmitted from the tag 110.
[0039] In step S220, the terminal device 400 sends an access request to the data management server 200 based on the access information received from the tag 110. In step S310, the data management server 200 receives the access request from the terminal device 400.
[0040] In step S315, the data management server 200 sends a password to the terminal device 400. From the viewpoint of enhancing security, it is preferable that the password is a one-time password that is valid for a certain period of time. In step S225, the terminal device 400 receives the password from the data management server 200.
[0041] In step S230, the terminal device 400 performs encryption processing. As an example of encryption processing, the terminal device 400 generates a hash value from the user ID it holds and the password received from the data management server 200. The terminal device 400 further encrypts the hash value based on location information obtained from GPS (Global Positioning System) or LTE (Long Term Evolution). As an encryption algorithm, for example, known post-quantum cryptographic algorithms such as lattice-based cryptography or hash-based cryptography may be used. As an example, the fluctuation value (decimal value) of the GPS value of the terminal device 400 may be used for lattice-based cryptography. Note that the location information is not limited to the examples above, and for example, MAC address or altitude may be used, or information that changes slightly over time but does not change significantly, such as radio wave strength or temperature, may be used.
[0042] In step S235, the terminal device 400 sends the generated encryption to the data management server 200. In step S320, the data management server 200 receives the encryption from the terminal device 400.
[0043] As shown in Figure 9, in step S330, the data management server 200 performs decryption processing of the received encrypted data. For example, the decryption process may involve checking the hash value generated by the user ID and password, decrypting the data using the geographic information of the registered tag 110, and authenticating the decrypted information using the location information obtained from the access IP (Internet Protocol Address) address of the terminal device 400. By implementing this process flow of encryption, decryption, and authentication using the terminal device 400, hacking (unauthorized access) from a different region than the equipment to which the tag 110 is attached can be prevented.
[0044] In step S335, the data management server 200 sends the data stored in the virtual storage 230 to the terminal device 400. In step S240, the terminal device 400 receives the data sent from the data management server 200. As a result, the user operating the terminal device 400 retrieves the data stored in the virtual storage 230 within the data management server 200.
[0045] (1.3. Data analysis server 300) (1.3.1. Functional configuration of data analysis server 300) Referring to Figure 10, the functional configuration of the data analysis server 300 will be described. The data analysis server 300 comprises a control unit 310 and a storage unit 350. The control unit 310 controls the processing of the data analysis server 300 as a processor. The control unit 310 comprises a fault detection module 320, a prediction module 330, and an optimization module 340.
[0046] The fault detection module 320 detects a fault in the equipment based on data detected by the detection unit and estimates the cause of the fault. Details of the processing performed by the fault detection module 320 will be described later.
[0047] The prediction module 330 predicts future power generation and power consumption at a facility that includes power generation equipment. For example, the prediction module 330 predicts power generation based on historical data of power generation and power consumption at the facility and weather data related to the location where the equipment is installed. For example, if the equipment is a solar panel P, the prediction module 330 predicts the future power generation by the solar panel P, the power consumption at the facility where the solar panel P is installed, and so on. Details of the processing performed by the prediction module 330 will be described later.
[0048] The optimization module 340 estimates the optimal operating policy for the equipment based on the power generation and power consumption predicted by the prediction module 330. As an example, the optimization module 340 proposes the optimal operating policy for the equipment in the facility in the future based on data detected by the detection unit. Details of the processing performed by the optimization module 340 will be described later.
[0049] The memory unit 350 stores programs and data necessary for processing by the control unit 310, data obtained through processing by the control unit 310, and so on. As an example, the memory unit 350 includes IoT historical data 355, failure data 360, learning data 365, failure mode correct answer data 370, weather data 375, failure detection model 380, prediction model 390, and optimization model 395.
[0050] IoT historical data 355 is historical data acquired from the current sensor 120 and the surveillance camera 130, which act as detection units. Failure data 360 is historical data of IoT historical data 355 when equipment failures occurred in the past. Training data 365 is data detected in general failures that are independent of individual equipment or the region in which they are installed. Specifically, it may include surges caused by lightning strikes and output waveform data of the power conditioner when a circuit breaker trips. Failure mode correct data 370 is correct data regarding the correspondence between the cause of the failure and the failure event (failure mode).
[0051] Weather data 375 is weather data for the environment in which the equipment is installed. Weather data 375 may be obtained from, for example, the Japan Meteorological Agency, the Japan Meteorological Business Support Center, or private companies. If there is no weather data corresponding to the equipment installation location in the obtained data, weather data may be generated by considering the influence of wind direction calculated from surrounding weather information, and / or the influence of mountain ranges around the equipment installation location.
[0052] The fault detection model 380 is a pre-trained model used by the fault detection module 320 when machine learning is performed to estimate fault detection. The prediction model 390 is a pre-trained model used by the prediction module 330 when machine learning is performed to estimate fault detection. The optimization model 395 is a data model used when processing is performed by the optimization module 340.
[0053] (1.3.2. Details of Model 380 for fault detection) The fault detection model 380 includes an anomaly detection model 381, a waveform classification model 382, a fault classification model 383, a fault mode model 384, an explanatory variable model 385, and a fault detection model 386.
[0054] Anomaly detection model 381 is a trained model that has learned whether the data acquired by the detection unit is abnormal or not. Waveform classification model 382 is a trained model that has learned the correspondence between the waveform of the data acquired by the detection unit and the device that output the data. Fault classification model 383 is a trained model that has learned the correspondence between the waveform of the data acquired by the detection unit and the classification of faults (i.e., fault events).
[0055] The failure mode model 384 is a pre-trained model that has learned the correspondence between the cause of failure and the event of failure (failure mode). The explanatory variable model 385 is a pre-trained model that has learned the correspondence between the data acquired by the detection unit when a failure occurs and the explanatory variables with high correlation. The failure detection model 386 is a pre-trained model that has learned the correspondence between the data acquired by the detection unit when a failure occurs and the cause of the failure.
[0056] (1.3.3. Processing of fault detection module 320) Referring to Figure 12, the processing flow of the fault detection module 320 will be explained. In step S410, pre-training processing is performed. In pre-training processing, Anomaly detection learning Waveform classification learning • Failure classification learning • Failure Mode Learning • Learning explanatory variables • Fault detection learning The following machine learning process is performed. This will be explained in detail below. Note that known machine learning algorithms can be used in the following learning process. Specific algorithms include, for example, linear regression, logistic regression, support vector machines, k-nearest neighbors, decision trees, random forests, neural networks, and Naive Bayes. These algorithms should be selected appropriately depending on the learning content, data structure, and computational resources. Furthermore, the pre-training process may be scheduled to run at predetermined intervals (for example, once a month).
[0057] As shown in Figure 13, in anomaly detection learning, historical IoT data 355 is used as input data, data with a high frequency of occurrence is treated as normal data, and data with a low frequency of occurrence is treated as abnormal data. The system learns to classify (categorize) normal data and abnormal data, and generates an anomaly detection model 381. Alternatively, the input data may be subjected to range detection processing, as described later, before determining the frequency of occurrence.
[0058] As shown in Figure 14, in waveform classification learning, IoT historical data 355 and training data 365 are used as input data to learn the correspondence between the waveforms of the data contained in IoT historical data 355 and the devices that output that data, and a waveform classification model 382 is generated.
[0059] As shown in Figure 15, in failure classification learning, IoT historical data 355, training data 365, and failure mode ground truth data 370 are used as input data to learn the correspondence between failure events and the waveforms of the failure data included in the failure mode ground truth data 370, and a failure classification model 383 is generated.
[0060] As shown in Figure 16, in failure mode learning, the failure data 360 is first normalized. Since the failure data 360 includes data on failures registered by construction company personnel during inspection work, the wording of the failure may differ depending on the person in charge. In the normalization process, the wording of the failures contained in the failure data 360 is quantified using an LLM (Large Language Model), etc., and those with a high similarity to the contents of the existing failure list are corrected to match the contents of the failure list. Also, for those with a low similarity to the contents of the existing failure list, a new failure is named and added to the failure list. By normalizing the description of the failures in this way, noise in the input data for subsequent learning processes can be reduced. The normalized data is stored in the failure mode ground truth data 370.
[0061] Next, using the correct failure mode data 370 as input data, the system learns failure modes, which are the correspondence between the cause of a failure and the resulting event. For example, in the case of solar panel failures, the system learns the correspondence between the cause of the failure and the event that may result from it, such as lightning strike + silicon burning, rockfall + glass breakage, and animal damage + cable disconnection, and generates a failure mode model 384.
[0062] As shown in Figure 17, in explanatory variable learning, first, input variable extraction is performed using IoT historical data 355 and weather data 375 as input variable data, and failure mode ground truth data 370 as the target variable data. In input variable extraction, from among the various data contained in IoT historical data 355 and weather data 375, only data with a high correlation value with the failure mode ground truth data 370 is extracted as input variable 396.
[0063] In calculating correlation values during input variable extraction, for example, a correlation matrix may be calculated and a DNN (Deep Neural Network) model may be used. Specifically, for example, the correlation coefficient between the input variable data and the target variable data is calculated, and a correlation matrix is created. Here, by normalizing the correlation coefficient values (for example, within the range of 0 to 1), the influence of explanatory variables can be compared on a unified scale. Furthermore, explanatory variables whose correlation coefficients are below a predetermined threshold may be treated as a fixed value of 0 in the DNN input layer, and the corresponding nodes may be deleted to reduce subsequent learning processing. In this way, by reducing the dimensionality of the explanatory variables, learning can be focused on important variables, improving the learning speed of the DNN model and reducing computational costs. After that, the performance of the DNN model is evaluated, and thresholds and parameters are adjusted as needed. As a result, 396 input variables are extracted.
[0064] Subsequently, explanatory variable learning is performed using the extracted input variables 396 and the ground truth failure mode data 370. That is, the correlation between the input variables 396 and the ground truth failure mode data 370 is learned, and an explanatory variable model 385 is generated.
[0065] As shown in Figure 18, in fault detection learning, IoT historical data 355, fault mode ground truth data 370, and weather data 375 are used as input data to learn the correspondence between the data waveforms of IoT historical data 355 and the fault modes, and a fault detection model 386 is generated. In addition, the multidimensional distance of IoT historical data 355 is calculated using fault mode correlation data 397, which shows the correlation values between fault modes. Here, the fault mode correlation data 397 is data relating to the correlation between the input variables 396 and the fault mode ground truth data 370.
[0066] Returning to Figure 12, let's continue the explanation. In step S420, preprocessing is performed on the data contained in the input IoT historical data 355. The preprocessing includes: • Range detection ·Waveform classification inference This process is carried out. Alternatively, the system may be configured to acquire IoT historical data 355 at predetermined intervals (for example, every few minutes) and perform preprocessing (S420) and subsequent inference processing (S430).
[0067] Range detection involves determining which ranges define the waveform boundaries (i.e., the waveform's period or repetitions) within the target data's waveform. This process clarifies which time periods should be included in the analysis, thereby improving the accuracy of subsequent inference processes.
[0068] In waveform classification inference, the waveform of the target data is classified to determine which equipment within the facility produced the output. Waveform classification is performed using waveform classification model 382.
[0069] In step S430, inference processing is performed. In inference processing, Anomaly detection and inference ·Fault classification inference • Fault detection and inference • Failure mode inference This will be carried out. The following explanation will be given with reference to Figure 19.
[0070] As shown in Figure 19, the inference process (S430) can take either pre-processed IoT historical data 355 as input data or fault data 360 as input data. In the case where IoT historical data 355 is used as input data, anomaly detection inference is performed first.
[0071] In anomaly detection inference, the preprocessed IoT historical data 355 is used with an anomaly detection model 381 to determine whether or not anomalies exist. If an anomaly is determined to exist, explanatory variable filtering is performed on the IoT historical data 355.
[0072] In explanatory variable filtering, explanatory variable model 385 is used to extract data from historical IoT data 355 that correspond to input variables 396 (see Figure 17) that have a high correlation with the ground truth failure mode data 370. After explanatory variable filtering is performed on the historical IoT data 355 in this manner, failure classification inference and failure detection inference are performed.
[0073] Furthermore, in cases where failure data 360 is used as input data, data corresponding to input variables 396 that have a high correlation with the ground truth failure mode data 370 is extracted from the failure data 360. In this way, explanatory variable filtering is performed on the IoT historical data 355, and then failure classification inference and failure detection inference are performed.
[0074] In failure classification inference, the failure classification (i.e., the failure event) is estimated using the failure classification model 383 on historical IoT data 355 or failure data 360 that have undergone explanatory variable filtering. The failure classification is estimated with the corresponding probabilities of multiple failure events, such as a 70 percent probability for "cable breakage" and a 30 percent probability for "glass breakage."
[0075] In fault detection inference, the cause of a failure is estimated using the fault detection model 386 on historical IoT data 355 or fault data 360, which have been filtered by explanatory variables. The cause of the failure is estimated to include multiple causes, such as a 60 percent probability of "lightning strike" and a 40 percent probability of "earthquake," along with their corresponding probabilities. Once the failure event and its cause are estimated, failure mode inference is performed.
[0076] In failure mode inference, failure modes, which are combinations of failure events and causes, are estimated for IoT historical data 355 or failure data 360, where multiple failure events and causes have been estimated. In failure mode estimation, the failure mode is estimated based on the estimated failure classification and failure cause using the following procedure. 1. Calculate the probability of all possible combinations of failure events and causes (failure mode probability). 2. Identify the highest failure mode probability, Pm, among the calculated probabilities. 3. Determine if the failure mode is registered in the correct data 370. 4. Identify the registered maximum probability P, which is the highest failure mode probability among those registered. 5. Calculate the frequency of occurrence of events and causes. 6. Calculate the confidence level.
[0077] The following explanation is based on the specific example shown in Figure 20. In the example shown in Figure 20, the probability of four failure events A to D has been calculated by the previous failure classification inference. In addition, the probability of three failure causes I to III has been calculated by the previous failure detection inference. First, in step 1, the failure mode probability is calculated for all combinations of events A to D and causes I to III (i.e., 12 combinations of A I, A II, ..., D III). Next, in step 2, the maximum probability Pm, which is the highest failure mode probability among those calculated, is identified. In the example shown in Figure 20, the failure mode probability of 0.56 for the combination A I is identified as Pm.
[0078] Next, in step 3, it is determined whether the failure mode is registered in the correct failure mode data 370. In the example shown in Figure 20, it is determined that "A-i", "C-i", "C-ha", and "D-i" are not registered as failure modes. Next, in step 4, the maximum registration probability P, which is the highest failure mode probability among those that are registered, is identified. In the example shown in Figure 20, the failure mode probability of 0.42 for the combination B-i is identified as P.
[0079] Next, in step 5, the number of occurrences of each event and cause is calculated. In the example shown in Figure 20, the number of occurrences of each event and cause is calculated after subtracting the failure modes that are not registered in the correct failure mode data 370 from the total number of combinations of events and causes. Specifically, as shown in Figure 20, the number of occurrences of event A is 2 because failure mode "A-i" is not registered. The number of occurrences of event B is 3. The number of occurrences of event C is 1 because failure modes "C-i" and "C-ha" are not registered. The number of occurrences of event D is 2 because failure mode "D-i" is not registered. Similarly, the number of occurrences of causes I to C is calculated.
[0080] Next, in step 6, we calculate the reliability R. Reliability is an index that shows the reliability of an estimated failure mode while considering the possibility of unknown failures. As an example, reliability R is calculated using the following formula. R = P - Pd + Pc Pd (Decreased Confidence Term): Maximum Probability Pm - Maximum Registration Probability P Pc (Confidence Increase Term): Number of occurrences of the event corresponding to the maximum registered probability P / Total number of occurrences of all events
[0081] In this way, the greater the maximum probability Pm is compared to the registered maximum probability P, the higher the probability of an unknown failure, and the lower the confidence level. Also, the more frequently an event corresponding to the registered maximum probability P occurs, the higher the probability of that event, and the lower the confidence level. By calculating this confidence level R, it is possible to accurately estimate the combination of failure events and causes while taking into account the occurrence of unknown failures.
[0082] In the example shown in Figure 20, the confidence level R is calculated as follows: P=0.42 Pm = 0.56 Pd=0.14 Pc = 3 / 8 = 0.375 R = 0.42 - 0.14 + 0.375 = 0.655
[0083] The specific formulas for calculating the confidence reduction term Pd and the confidence increase term Pc are not limited to the examples above. For example, the confidence R may be calculated using the above formula after multiplying at least one of the confidence reduction term Pd or the confidence increase term Pc by a predetermined coefficient. In addition, the confidence increase term Pc may include (or replace) the occurrence rate of the cause.
[0084] The fault detection module 320 displays multiple fault modes (i.e., combinations of fault events and causes) with a high probability of failure on the terminal device. In this case, the module may also display the confidence level R along with the probability of the fault mode. Here, the confidence level R may be displayed by comparing the calculated confidence level R value with a predetermined reference value and indicating it as "high confidence," "medium confidence," "low confidence," etc.
[0085] (1.3.4. Processing of prediction module 330) Referring to Figures 21 to 25, the processing flow of the prediction module 330 and the optimization module 340 will be explained. As shown in Figure 21, in step S510, a pre-training process is performed. In the pre-training process, • Learning explanatory variables • Predictive learning This machine learning process is performed. The following explains it step by step.
[0086] As shown in Figure 22, in explanatory variable learning, IoT historical data 355 and weather data 375 are used as input variable data, and IoT historical data 355 is used as the target variable data to extract the input variables. The IoT historical data 355 in the input variable data includes actual power consumption data of the equipment. The IoT historical data 355 in the target variable data includes actual power generation data.
[0087] Subsequently, explanatory variable training is performed using the extracted input variables 396 and IoT historical data 355. That is, the correlation between the input variables 396 and IoT historical data 355 is learned, and an explanatory variable model 385 is generated.
[0088] As shown in Figure 23, in predictive learning, machine learning is performed using IoT historical data 355 and weather data 375 as input data to generate a predictive model 390. The IoT historical data 355 may include the amount of power generated by solar panels P, power consumption, equipment information, inspection information, repair information, etc. The algorithm used in the fault detection model 380 described above may be used for machine learning.
[0089] Furthermore, the explanatory variable model 385 described above may be used when training the prediction model 390. Specifically, instead of the failure mode ground truth data 370, data on the power generation amount of solar panels P included in the IoT historical data 355 may be used to extract input variables from the IoT historical data 355 and weather data, using data with a high correlation to power generation as input variables. Subsequently, predictive learning may be performed using the extracted input variables and predictor variables that include data on the power generation amount of solar panels P included in the IoT historical data 355.
[0090] Next, in step S520, predictive inference is performed. In the example shown in Figure 24, as an example of predictive inference, IoT historical data 355 and weather data 375 are used as input data, and after filtering of explanatory variables, the amount of power generated by the solar panel P is estimated using the predictive model 390. The predictive inference may be performed at a predetermined frequency (for example, once a day). Furthermore, the estimated amount of power generated may be corrected to account for the aging degradation of the solar panel P. In addition, the parameters of the predictive model 390 may be adjusted as appropriate to improve the prediction accuracy by comparing the estimated predicted amount with the actual amount of power generated, which is updated daily.
[0091] (1.3.5. Processing of Optimization Module 340) The optimization process performed by the optimization module 340 will be explained with reference to Figure 25. The optimization module 340 estimates the optimal operating policy for the power consumption of equipment in a facility that includes solar panels P. As an example, in the example shown in Figure 21, the optimal operating policy is estimated for a house H where solar panels P are installed, based on the amount of charge in the rechargeable battery B and the amount of power consumed by the home appliance E.
[0092] (Optimization under normal conditions) First, let's discuss the optimization of power during normal operation. In this example, rechargeable battery B can be charged by solar panels P and by a wall outlet. In this case, • Input power from solar panel P: P1(t) • Power input from the outlet: P2(t) • Power output for consumption by home appliances: P3(t) ·Natural discharge amount: P4 ·Charging amount: P5(t) ·Waste electricity amount: P6(t) ·Power consumption: P7(t) Battery level: Pbm(t) This leads to an optimization problem where we want to reduce electricity costs (i.e., decrease the input energy P2) while adjusting these eight parameters. It is assumed that natural discharge is constant regardless of time t.
[0093] The optimization module 340 calculates an approximate solution to the above optimization problem using quantum annealing. Specifically, first, The Hamiltonian Hmin is set for the objective function JPEG2026182272000002.jpg12147. Here, Hcom is the Hamiltonian of the constraints. The weight coefficients λ1 and λ2 are subject to the condition λ1 < λ2, such that the input energy P2(t) from the outlet is less than the waste energy P6(t). Also, t=0 and n are arbitrary times; for example, t=0 may be 24 hours before the present and t=n may be 24 hours after the present.
[0094] The constraints may be, for example, as follows: JPEG2026182272000003.jpg4147 Here, He: Conditions related to energy balance Hbat: Conditions for updating battery level Hre: Conditions regarding the non-negativity of battery charge Hch: Conditions regarding the prohibition of simultaneous charging and discharging. Had: Priority criteria for power-consuming home appliances That is the case.
[0095] Specifically, since the input power to rechargeable battery B and the input power from rechargeable battery B are equal, the equation P1(t)+P2(t)+Pbm(t)=P3(t)+P4(t)+P5(t) holds true. Therefore, The file name will be JPEG2026182272000004.jpg12147.
[0096] Furthermore, focusing on the remaining charge of rechargeable battery B, the equation Pbm(t+1)=Pbm(t)+P5(t)-P7(t)-P4 holds true. Therefore, The file name will be JPEG2026182272000005.jpg12147.
[0097] Furthermore, since the remaining charge of rechargeable battery B cannot be negative, Pbm(t)≧0 holds true. Therefore, The file name will be JPEG2026182272000006.jpg12147.
[0098] Furthermore, since charging and discharging of rechargeable battery B cannot occur simultaneously, the equation P5(t) × P7(t) = 0 holds true. Therefore, The file name will be JPEG2026182272000007.jpg12147.
[0099] Furthermore, considering the power consumption priority of each home appliance, The result is JPEG2026182272000008.jpg15147. Here, δ(i) is an adjustment coefficient assigned to each of the multiple (1 to m) home appliances. The higher the priority of the home appliance, the larger the adjustment coefficient δ(i). Note that under normal circumstances, the adjustment coefficient may be kept constant. In addition, the weighting coefficient λ3 is subject to the condition λ3 < λ2, such that the amount of output power P3(t) consumed by the home appliance is less than the amount of waste power P6(t).
[0100] In this way, the optimization module 340 can set the Hamiltonian Hcom of the constraints, calculate an approximate solution of the objective function Hmin by quantum annealing, and estimate Pn(t) (where n=1 to 7). In this way, in this embodiment, for facilities including power generation equipment, it is possible to estimate the optimal operating policy for each facility while considering the external environment and internal power consumption of the facility.
[0101] Each set Hamiltonian is stored in the optimization model 395. Existing quantum computing services, such as those of D-Wave Systems, may be used to perform the quantum annealing. Furthermore, the optimization module 340 may be implemented using a gate-type quantum computer instead of quantum annealing, or an approximate solution to the optimization problem may be calculated using a simulated annealing method that does not use a quantum computer. While the optimization calculates an approximate solution that minimizes the input power P2 from the outlet, negative values for the input power P2 may be permitted. In this way, the technical concept of this disclosure can be applied even in cases where the generated electricity is sold externally.
[0102] (Optimization during disasters) As a disaster scenario, we assume a case where power supply is cut off after 6 hours due to planned blackouts, etc. In this case, the constraint Hcom is: The image may also be named JPEG2026182272000009.jpg21147. Here, to=6, and P2(t)=0 for t≧to. By setting the Hamiltonian Hcom with such constraints, it is possible to estimate the optimal power supply operation policy for each facility in the event of an interruption in the power supply, such as during a disaster.
[0103] In the event of a disaster, the adjustment coefficient δ for home appliances is adjusted according to priority in equation (7). Specifically, for example, the adjustment coefficient δ(i) > 0 for home appliances with high priority during a disaster, such as refrigerators and communication equipment, and the adjustment coefficient δ(i) = 0 for other home appliances. In this way, the power consumption of the equipment can be set considering the priority of home appliances, and the future operating policy of the equipment can be determined. In addition, explanatory variable learning may be used in adjusting the adjustment coefficient δ. That is, by learning the correlation between the actual values of multiple output energy quantities P3 and the actual values of input energy quantities P2, it may be possible to identify home appliances that contribute significantly to the input from the outlet and increase the priority of those appliances (i.e., increase the adjustment coefficient δ).
[0104] (1.4. Hardware configuration of the information processing device) Referring to Figure 26, the hardware configuration of an information processing device used as a data management server 200, a data analysis server 300, a terminal device 400, etc., will be described. As an example, the information processing device is implemented by the computer 90 shown in Figure 26. The computer 90 may include a CPU 91, ROM 92, RAM 93, storage 94, an input interface 95, an output interface 96, and a communication interface 97.
[0105] The CPU 91 functions as a processor that executes processing. Specifically, the CPU 91 uses the RAM 93 as work memory and executes programs stored in at least one of the ROM 92 or storage 94. During program execution, the CPU 91 controls each component via the system bus 98 and performs various processes. For example, the CPU 91 functions as a control unit 210 or a control unit 310.
[0106] ROM92 stores programs that control the operation of computer 90. ROM92 contains the programs necessary for computer 90 to perform each of the processes described above. RAM93 functions as a memory area where the programs stored in ROM92 are loaded.
[0107] The storage 94 stores data necessary for program execution and data obtained through program execution. The storage 94 includes one or more selected from Hard Disk Drives (HDDs) and Solid State Drives (SSDs). As an example, the storage 94 functions as a memory unit 220.
[0108] The input interface (I / F) 95 can connect the computer 90 and the input device 95a. The input interface 95 is, for example, a serial bus interface such as USB. The CPU 91 can read various data from the input device 95a via the input interface 95.
[0109] The output interface (I / F) 96 can connect the computer 90 to the output device 96a. The output interface 96 is a video output interface such as Digital Visual Interface (DVI) or High-Definition Multimedia Interface (HDMI®). The CPU 91 can transmit data to the output device 96a via the output interface 96 and cause the output device 96a to output data.
[0110] The input device 95a is an example of an input means and includes one or more selected from a mouse, keyboard, microphone (voice input), and touchpad. The output device 96a is an example of an output means and includes one or more selected from a display, projector, printer, and speaker. Devices that have the functions of both the input device 95a and the output device 96a, such as a touch panel, may also be used.
[0111] The communication interface (I / F) 97 allows the computer 90 to connect with an external server 97a located outside the computer 90. The communication interface 97 is, for example, a network card such as a LAN card. The CPU 91 can read various data from the external server 97a via the communication interface 97.
[0112] Each process performed by the data management server 200, the data analysis server 300, or the terminal device 400 may be implemented by a single computer 90, or by the collaborative efforts of multiple computers 90.
[0113] The processing of the various data described above may be recorded as a program that can be executed by a computer on a magnetic disk (flexible disk and hard disk, etc.), an optical disk (CD-ROM, CD-R, CD-RW, DVD-ROM, DVD±R, DVD±RW, etc.), a semiconductor memory, or another non-transitory computer-readable storage medium.
[0114] For example, information recorded on a recording medium can be read by a computer (or embedded system). The recording format (storage format) of the recording medium is arbitrary. For example, a computer reads a program from the recording medium and, based on this program, causes the processor to execute the instructions written in the program. In a computer, program acquisition (or reading) may be performed via a network.
[0115] <2. Other Embodiments> The maintenance management system 1 according to this embodiment has been described above, but the application of the technical ideas of this disclosure is not limited to the above embodiment. For example, satellite images acquired from an artificial satellite S may be used for fault detection. Specifically, for example, images from a predetermined time ago (e.g., 30 minutes ago) and the present are acquired from satellite images, and by performing difference analysis and motion analysis, changes and moving objects in the images are extracted. Furthermore, in conjunction with time and GPS information, a database of similar images and fault cases taken in the past is referred to and compared with the current situation to detect abnormalities. For example, a person who breaks in at night with the intention of theft will show a different time and movement pattern than a repair worker, so such unique characteristics can be detected.
[0116] While several embodiments of this disclosure have been illustrated above, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made 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 of the invention and its equivalents. Furthermore, the embodiments described above can be implemented in combination with each other. [Explanation of symbols]
[0117] 1: Maintenance management system, 90: Computer, 91: CPU, 92: ROM, 93: RAM, 94: Storage, 95: Input interface, 95a: Input device, 96: Output interface, 96a: Output device, 97: Communication interface, 97a: External server, 98: System bus, 100: IoT device, 110: Tag, 120: Current sensor, 130: Surveillance camera, 200: Data management server, 210: Control unit, 220: Memory unit, 230: Virtual storage, 230a: IoT data management table, 30 0: Data analysis server, 310: Control unit, 320: Fault detection module, 330: Prediction module, 340: Optimization module, 350: Memory unit, 355: IoT historical data, 360: Fault data, 365: Training data, 370: Ground truth failure mode data, 375: Weather data, 380: Fault detection model, 381: Anomaly detection model, 382: Waveform classification model, 383: Fault classification model, 384: Fault mode model, 385: Explanatory variable model, 386: Fault detection model, 390: Prediction model, 395: Optimization module Dell, 396: Input Variables, 397: Failure Mode Correlation Data, 400: Terminal Device, 410: Management Screen, 411: Menu List, 411a: Equipment List Icon, 412: Organization Page, 413: User Information, 414: Organization Information Dropdown List, 414a: Organization Management Icon, 420: Equipment List Screen, 421: Equipment List, 422: Equipment Details Button, 423: Application Edit Button, 430: Equipment Details Screen, 431: Equipment Details, 432: Equipment Details Edit Button, 440: Equipment Details Edit Screen, 441: Display Settings List, 4 42: Public setting checkbox, 450: Usage management screen, 451: Usage list, 452: Details button, 460: Usage details screen, 461: Usage item list, 462: Edit button, 470: Details editing screen, 471: Item details, 472: Guest hide checkbox, 480: Organization management screen, 481: User list, 482: Details button, 490: User information details screen, 491: User information details list, 492: User information edit button, 495: User information edit screen, 496: Inspection registration permission checkbox
Claims
1. A prediction module that predicts future power generation and power consumption in a facility including power generation equipment, The system includes an optimization module that estimates the optimal operating policy for the equipment in the facility based on the predicted amount of power generated and the amount of power consumed, The prediction module is a system that predicts the amount of power generated based on actual power generation and power consumption data at the facility and weather data related to the location where the equipment is installed.
2. The aforementioned system, The system further includes a detection unit that detects data relating to the status of the equipment or the status of the environment in which the equipment is installed. The system according to claim 1, wherein the detection unit detects the actual data of power generation and power consumption.
3. The system according to claim 2, wherein the prediction module uses the data detected by the detection unit and the weather data as input variables, performs machine learning with the amount of power generated as the target variable, and generates a trained model for predicting the amount of power generated.
4. The system according to claim 3, wherein the prediction module predicts the amount of power generation by extracting parameters that have a high correlation with the amount of power generation from the data detected by the detection unit and the weather data, and inputting the extracted parameters into the trained model.
5. It also has a battery to charge the generated electricity, The system according to claim 1, wherein the optimization module calculates an approximate solution to a minimization problem that minimizes an objective function whose parameters include the amount of power generated, the amount of power input from the outlet, the amount of power consumed, the amount of charge of the battery, and the amount of waste power, when estimating the optimal operating policy.
6. The optimization module uses a quantum computer to calculate the approximate solution, as described in claim 5.
7. The system according to claim 5, wherein the optimization module extracts parameters that are highly correlated with the input power from among a plurality of parameters relating to the power consumption, and calculates the approximate solution.
8. Equipped with a processor and memory, The aforementioned processor, A prediction module that predicts future power generation and power consumption in a facility including power generation equipment, It includes an optimization module that estimates the optimal operating policy for the equipment in the facility based on the predicted amount of power generated and the amount of power consumed, The prediction module is an information processing device that predicts the amount of power generated based on actual power generation and power consumption data at the facility and weather data related to the location where the equipment is installed.
9. A program for operating an information processing device comprising a processor and memory, The aforementioned processor, A step to predict future power generation and power consumption at a facility including power generation equipment, The system is then configured to perform the following steps: estimate the optimal operating policy for the equipment in the facility based on the predicted amount of power generated and the amount of power consumed; The program, in the step of making a prediction, predicts the amount of power generated based on actual power generation and power consumption data at the facility and weather data related to the location where the equipment is installed.
10. An information processing method for operating an information processing device comprising a processor and memory, The aforementioned processor, A step to predict future power generation and power consumption at a facility including power generation equipment, The system is then configured to perform the following steps: estimate the optimal operating policy for the equipment in the facility based on the predicted amount of power generated and the amount of power consumed; An information processing method that, in the step of making a prediction, predicts the amount of power generated based on actual power generation and power consumption data at the facility and weather data relating to the location where the equipment is installed.
Citation Information
Patent Citations
Maintenance management system and maintenance management method
JP2021086237A
