Accident occurrence probability estimation device, control method and program for accident occurrence probability estimation device

Abstract

To provide an accident occurrence probability estimation device for obtaining probability of an accident occurrence of a distribution system in a predetermined area within a certain period of time, and a control method and a program thereof.SOLUTION: In an accident occurrence probability estimation system, an accident occurrence probability estimation device includes: a learning model storage unit 101 for storing a learning model prepared so as to calculate probability of an accident occurrence of a distribution system in an area within a certain period of time, when prediction information of an event in a certain period of time designating an area is input, based on factor information indicating presence / absence of an accident of a distribution system in the past in a plurality of areas and a status of a predetermined event in each area which may be a cause of an accident; a prediction information receipt unit 102 for receiving input of the prediction information designating the area; an accident occurrence probability acquisition unit 103 for acquiring the probability of the accident occurrence in a distribution system within a certain period of time in the area by inputting the area and the prediction information to the learning model; and an accident occurrence probability output unit for 104 outputting the probability of the accident occurrence.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to an accident occurrence probability estimation device, a control method for an accident occurrence probability estimation device, and a program. [Background technology]

[0002] In recent years, technological advances in power distribution systems have reduced the frequency of power outages, but in bad weather and other times, power outages can occur due to accidents such as broken or short-circuited power lines and ground faults.

[0003] For this reason, when a typhoon is approaching, for example, electric power companies prepare for accidents by securing workers and work equipment in advance.

[0004] However, it is difficult to predict when and where such a power distribution system accident will occur.

[0005] In connection with the prediction of such faults in power distribution systems, techniques for evaluating the reliability of power distribution systems have been developed (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2005-094870 Summary of the Invention [Problem to be solved by the invention]

[0007] However, the prediction of such power distribution system accidents is influenced by the past experience of the person in charge, and for example, it is difficult for inexperienced personnel to predict accidents. Furthermore, even experienced personnel may have different perspectives on situation analysis, and the results of accident predictions may differ depending on the person in charge.

[0008] The present invention has been made in view of the above background, and aims to provide an accident probability estimation device, a control method for an accident probability estimation device, and a program that can calculate the probability of an accident occurring in a power distribution system in a specified area within a certain period of time. [Means for solving the problem]

[0009] One of the present inventions for achieving the above object is an accident occurrence probability estimation device for calculating the probability of an accident occurring in a distribution system in a predetermined area within a certain period of time, the device comprising: a learning model storage unit that stores a learning model created to calculate the probability of an accident occurring in the distribution system in a specified area within a certain period of time when prediction information of an event for a specified area within the certain period of time is input based on whether or not there have been any past accidents in the distribution system in a plurality of areas including the specified area and factor information indicating the status of a specified event in each area that may be the cause of an accident; a prediction information receiving unit that receives input of the prediction information for a specified area; an accident occurrence probability acquisition unit that acquires the probability of an accident occurring in the distribution system in the area within the certain period of time by inputting the area and the prediction information into the learning model; and an accident occurrence probability output unit that outputs the probability of an accident occurring. The certain period includes a plurality of periods of different lengths, and the accident occurrence probability acquisition unit acquires the probability of an accident occurring for each day within the certain period from the learning model, and calculates a value obtained by subtracting the product of the probability that an accident will not occur each day from 1 for each period, thereby calculating the probability of an accident occurring for each period. do.

[0010] Other problems and solutions disclosed in the present application will be made clear in the detailed description and drawings. [Effects of the Invention]

[0011] According to the present invention, it is possible to obtain the probability that an accident will occur in a power distribution system in a predetermined area within a certain period of time. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a diagram illustrating an accident occurrence probability estimation system. [Figure 2] FIG. 2 is a hardware configuration diagram of the accident occurrence probability estimation device. [Figure 3]FIG. 2 is a diagram illustrating a storage device of the accident occurrence probability estimation device. [Figure 4] FIG. 10 illustrates a cause information management table. [Figure 5] FIG. 1 is a diagram illustrating a learning model. [Figure 6] FIG. 2 is a functional block diagram of the accident occurrence probability estimation device. [Figure 7] 3 is a flowchart showing the flow of processing by the accident occurrence probability estimation device. [Figure 8] 3 is a flowchart showing the flow of processing by the accident occurrence probability estimation device. [Figure 9] FIG. 10 is a diagram showing an example of a display screen showing the probability of an accident occurring. DETAILED DESCRIPTION OF THE INVENTION

[0013] At least the following matters will become apparent from the description of this specification and the accompanying drawings. Hereinafter, the present invention will be described in accordance with one embodiment thereof with reference to the accompanying drawings.

[0014] ==Overall Configuration== FIG. 1 shows the overall configuration of an accident occurrence probability estimation system 1000 according to an embodiment of the present invention.

[0015] The accident occurrence probability estimation system 1000 is configured by connecting an accident occurrence probability estimation device 100 and a power distribution line automation system 200 so that they can communicate with each other via a network 500 such as the Internet, a LAN (Local Area Network), or a telephone network. The accident occurrence probability estimation device 100 is also connected to a weather information providing device 300 so that they can communicate with each other via the network 500.

[0016] The power distribution line automation system 200 is a computer system that monitors and controls devices such as switches and transformers installed in a power distribution system, collects measurement data such as voltage and current, and detects and recovers from faults.

[0017] The power distribution system is divided into multiple sections by switches and managed, and is constructed across multiple areas. In this embodiment, the areas correspond to the sections of the power distribution system, but they do not necessarily have to correspond to the sections. For example, the areas may be defined based on the sections of the power distribution system, such that one area is composed of one or more sections. Alternatively, the areas may refer to administrative districts such as cities, wards, towns, and villages. If the areas do not correspond to the sections, a table (not shown) that associates the areas with the sections is prepared in the accident probability estimation device 100.

[0018] The distribution line automation system 200 identifies the type of fault, such as a short circuit, ground fault, or disconnection, and the section where the fault occurred, based on the open / close state of each switch and measurement data such as voltage and current acquired from various sensors installed in the power distribution system.The distribution line automation system 200 then controls the switches to isolate the section where the fault occurred from the other sections and cause a power outage, while allowing power to be supplied from the substation to healthy sections where no fault has occurred.

[0019] When the distribution line automation system 200 detects such an accident in the distribution system, it transmits accident information including the date and time of the accident, the section where the accident occurred, the type of accident (short circuit, ground fault, open circuit, etc.) to the accident occurrence probability estimation device 100.

[0020] The accident occurrence probability estimation device 100 is a computer capable of calculating the probability that an accident will occur in a power distribution system within a certain period (for example, one week) in a predetermined area (for example, section A of a power distribution system).

[0021] The accident probability estimation device 100 stores the history of the above-mentioned accident information and the history of factor information indicating the situation in each area of ​​a specified event, such as weather information that may cause an accident in the power distribution system, in a factor information management table 600 described below.

[0022] The accident probability estimation device 100 acquires the section where the accident occurred from the power distribution line automation system 200, and as described above, in this embodiment, the area means the section. If the area and the section do not match, the accident probability estimation device 100 converts the section into an area by referring to the table that associates the above-mentioned areas with the sections, and stores the area in the cause information management table 600.

[0023] The accident occurrence probability estimation device 100 acquires accident information from the power distribution line automation system 200 and weather information from the weather information providing device 300, but acquires other event information from measuring instruments (not shown) that detect the status of each event, a computer (not shown), etc. The accident occurrence probability estimation device 100 also records cause information in the cause information management table 600 in association with the date for days when no accidents occurred in the power distribution system.

[0024] The accident occurrence probability estimation device 100 then analyzes this information to generate a learning model 620 (AI model) described below. When prediction information for the above-mentioned event specifying an area (e.g., section A of a power distribution system) within a certain period (e.g., each day within one week from today) is input, this learning model 620 outputs the probability of an accident occurring in the power distribution system in this area within the certain period. By using this learning model 620, when prediction information for the event specifying an area within the certain period is input, the accident occurrence probability estimation device 100 calculates and outputs the probability of an accident occurring in the power distribution system in this area within the certain period.

[0025] This method makes it possible to calculate the probability that a distribution system accident will occur in a specified area within a certain period of time. This reduces the variability in accident predictions that arise due to differences in the experience of distribution system managers and differences in the perspectives of situation analysis. It also makes it possible to more accurately formulate accident response plans, prepare personnel and equipment necessary to respond to accidents, or update distribution system equipment.

[0026] Furthermore, based on the output results, the person in charge can consider the following countermeasures. For example, if the probability of a power distribution line accident in the target area within one week is above a certain level (a standard such as 70% or above can be set in advance), it is possible to prevent power distribution line accidents before they occur, such as by conducting emergency patrols. Also, if the probability of a power distribution line accident in the target area within one day is above a certain level (a standard such as 50% or above can be set in advance), it is possible to take countermeasures such as securing recovery response personnel in advance.

[0027] The weather information providing device 300 is a computer that provides weather information such as weather (sunny, rainy, cloudy, typhoon, snow, hail, etc.), temperature, humidity, precipitation probability, precipitation amount, lightning probability, wind speed, and wind direction.

[0028] The accident probability estimation device 100 acquires predetermined types of weather information such as the weather and temperature of each area every day from the weather information providing device 300 and stores the acquired information in the factor information management table 600 as factor information in association with date information.

[0029] In this way, by including meteorological information related to power distribution system accidents in the factor information, the accuracy of the learning model is improved, and the accident probability estimation device 100 is able to output a more accurate probability of an accident occurring.

[0030] Details will be provided below.

[0031] ==Accident Probability Estimation Device== Next, the accident occurrence probability estimation device 100 will be described.

[0032] 2 shows an example of the hardware configuration of the accident occurrence probability estimation device 100. The accident occurrence probability estimation device 100 is configured to include a CPU (Central Processing Unit) 110, a memory 120, a communication device 130, a storage device 140, an input device 150, an output device 160, and a recording medium reading device 170.

[0033] The CPU 110 is responsible for overall control of the accident occurrence probability estimation device 100, and realizes various functions of the accident occurrence probability estimation device 100 by reading out the accident occurrence probability estimation device control program 700, which is composed of codes for performing various operations related to this embodiment stored in the storage device 140, and various data into the memory 120 and executing or processing them.

[0034] For example, the CPU 110 executes or processes the accident probability estimation device control program 700 and various data, and by working in cooperation with hardware devices such as the memory 120, communication device 130, and storage device 140, various functions such as the learning model memory unit 101, prediction information receiving unit 102, accident probability acquisition unit 103, and accident probability output unit 104, which will be described later, are realized.

[0035] The accident probability estimation device control program 700 is a general term for programs for realizing the functions of the accident probability estimation device 100, and includes, for example, application programs, an OS (Operating System), various libraries, etc. that run on the accident probability estimation device 100.

[0036] The memory 120 can be configured, for example, by a semiconductor memory device.

[0037] The storage device 140 is a device that provides a physical storage area for storing various programs, data, tables, etc., such as a hard disk drive, SSD (Solid State Drive), flash memory, etc. In this embodiment, as shown in Fig. 3, the storage device 140 stores various data such as a cause information management table 600, learning data 610, learning model 620, and estimation results 630 in addition to an accident occurrence probability estimation device control program 700.

[0038] An example of the cause information management table 600 is shown in FIG.

[0039] When an accident occurs in the power distribution system and accident information including the date and time of the accident and the section where the accident occurred is transmitted from the power distribution line automation system 200, the accident occurrence probability estimation device 100 stores this accident information in the cause information management table 600. In addition, when the accident type (disconnection, short circuit, ground fault, etc.) is acquired from the power distribution line automation system 200, the accident occurrence probability estimation device 100 also stores it in the cause information management table 600.

[0040] The accident probability estimation device 100 acquires predetermined types of daily weather information for each area from the weather information providing device 300 and stores it in the cause information management table 600. For example, the types of weather information include things that can cause accidents in the power distribution system, such as weather (sunny, cloudy, rain, snow, thunder, hail, etc.), temperature, humidity, precipitation, wind speed, wind direction, and atmospheric pressure.

[0041] In the factor information management table 600 shown in FIG. 4, the information listed in the "Type" and "Section" columns is information acquired from the power distribution line automation system 200, and the information listed in the "Weather Information" column is information acquired from a weather information providing device.

[0042] In addition, the information listed in the "Inspection Results" column is information regarding the inspection results of the distribution system equipment installed in each area (section), and the information listed in the "Usage Period" is information indicating the usage period of the distribution system equipment installed in each area (section).

[0043] A power distribution system is equipped with various pieces of equipment, such as pole transformers, insulators, cross arms, and switches, and maintenance such as inspections and replacements are performed periodically by electric power companies and their partner companies. The electric power companies record and manage information such as the inspection results of this equipment and the dates of installation and replacement in a computer (not shown). The accident probability estimation device 100 obtains this information from the computer and records it as cause information in the "inspection results" and "usage period" columns of the cause information management table 600.

[0044] By including this information in the factor information, the accuracy of the learning model 620 is improved, and the accident occurrence probability estimation device 100 is able to output a more accurate accident occurrence probability.

[0045] The information listed in the "Seasonal Factors" column indicates seasonally specific events that occur in the area and the times when those events are likely to occur, and the information listed in the "Regional Factors" column is information about geographically specific events that exist in the area.

[0046] For example, one area may be located along a windy coast and tend to have an increased number of short-circuit accidents in power lines during strong winds in bad weather, another area may be home to many wild birds that build nests on utility poles to raise their young during a specific period each year (e.g., March to July), and there may also be areas where the number of accidents involving snakes shorting out power lines increases during certain periods (e.g., April to October).

[0047] Information on potential accident causes specific to such areas and seasons is recorded in a computer (not shown), and the accident probability estimation device 100 can acquire this information from the computer and record it as factor information in the "seasonal factors" and "regional factors" columns of the factor information management table 600.

[0048] In this way, by including this information in the factor information, the accuracy of the learning model 620 is further improved, and the accident occurrence probability estimation device 100 is able to output a more accurate accident occurrence probability.

[0049] The "Cause of Accident" column contains the true cause of the accident that was discovered during an investigation after the accident. These causes of the accident are recorded in the cause information management table 600 by, for example, a person in charge inputting the cause of the accident via the input device 150.

[0050] 4, information for each area (section) is recorded together in a single table, but separate tables may be used for each area. The factor information management table 600 also records the date and factor information for days on which no accidents occurred. Therefore, if the type of accident or the accident section is recorded in the "Type" and "Section" columns in the factor information management table 600, where accident information is recorded, this means that an accident occurred on that day; if no information is recorded, this means that no accident occurred on that day. In other words, the accident probability estimation device 100 can obtain the presence or absence of accidents each day in each area by referencing the accident information in the factor information management table 600.

[0051] Returning to FIG. 3, the accident occurrence probability estimation device 100 generates learning data 610 based on the presence or absence of accidents (accident information) in each area (section) on each day during a predetermined past period (one month, six months, one year, etc., hereinafter referred to as a "time section") recorded in the factor information management table 600, and the factor information in each area on each day during the time section.

[0052] By using this learning data 610 to train a learning model 620 (machine learning model), when an area is specified and prediction information on factor information for each day within a certain period (for example, one week from the current day) is input, the learning model 620 can calculate the probability of a distribution system accident occurring in that area within that period.

[0053] The accident occurrence probability estimation device 100 generates learning data 610 by using factor information for each area on each day in the time interval as an explanatory variable (feature) and the presence or absence of an accident (accident information) on each day within the time interval as a target variable (label).

[0054] The time interval is set empirically based on, for example, the frequency of accidents. Alternatively, a learning model 620 may be generated for each of a plurality of time intervals, and the probability of an accident occurring in the power distribution system may be calculated using these learning models 620. The time interval is set to a length that improves the accuracy of the accident cause, for example, based on past cases.

[0055] Alternatively, the accident probability estimation device 100 may use data from a specific area to generate the training data 610. In this manner, a training model 620 specific to the area can be generated, thereby improving the accuracy of the accident probability.

[0056] Similarly, the accident occurrence probability estimation device 100 may use data from a specific period (e.g., a specific season or a specific month) to generate the training data 610. In this manner, a training model 620 specific to that period can be generated, thereby improving the accuracy of the accident occurrence probability.

[0057] Furthermore, the accident probability estimation device 100 may use data for a specific area and time period to generate the training data 610. In this manner, a training model 620 specific to the area and time period can be generated, thereby improving the accuracy of the accident probability.

[0058] In this embodiment, the learning model 620 is a DNN (Deep Neural Network), but may be realized by other types of models such as gradient boosting (GBDT (Gradient Boosting Decision Tree)).

[0059] FIG. 5 shows an example of a learning model 620 (neural network structure). As shown in FIG. 5, factor information for each area on each day within a time period is input to an input layer 621 of the learning model 620. The intermediate layer 622 includes one or more hidden layers each consisting of one or more nodes containing parameters adjusted by learning. Based on the factor information provided to the input layer 621, the intermediate layer 622 calculates one or more predicted values ​​(probability of an accident occurring) for the output layer 623. The output layer 623 outputs one or more predicted values ​​(probability of an accident occurring).

[0060] Returning to FIG. 3 , the estimation result 630 includes information indicating the probability of an accident occurring, output from the learning model 620. The learning model 620 according to this embodiment outputs the probability of an accident occurring over multiple time periods, such as the probability of an accident occurring within one day and the probability of an accident occurring within one week. This configuration can reduce variations in accident predictions that arise due to differences in the experience of distribution system managers and the perspectives of situation analysis. It also enables more accurate formulation of accident response plans, preparation of personnel and equipment required for accident response, and updating of distribution system facilities.

[0061] Returning to FIG. 2, the storage device 140 may be built into the accident occurrence probability estimation device 100 or may be externally attached.

[0062] The recording medium reader 170 reads programs and data recorded on a recording medium 800 such as a CD-ROM or DVD, and stores them in the storage device 140 .

[0063] The communication device 130 transmits and receives data and programs to and from other computers such as the power distribution line automation system 200 and the weather information providing device 300 via a network 500 such as the Internet or a LAN (Local Area Network). For example, the above-described accident occurrence probability estimation device control program 700 can be stored in another computer (not shown), and the accident occurrence probability estimation device 100 can download and execute the accident occurrence probability estimation device control program 700 from this computer.

[0064] Alternatively, the communication device 130 may periodically receive accident information and weather information from the power distribution line automation system 200 and the weather information providing device 300. Furthermore, the accident occurrence probability estimation device 100 may not include the storage device 140, and may instead realize its functions by using various data such as the above programs and tables stored in another computer (not shown) communicably connected via the network 500.

[0065] The input device 150 functions as a user interface and is used by a person in charge or the like to input data to the accident occurrence probability estimation device 100. The input device 150 may be, for example, a keyboard, a mouse, a microphone, or the like.

[0066] The output device 160 is a device for outputting information to the outside and functions as a user interface. The output device 160 may be, for example, a display, a printer, a speaker, or the like.

[0067] <Functional configuration> 6 shows a functional block diagram of the accident probability estimation device 100 according to this embodiment. The accident probability estimation device 100 includes a learning model storage unit 101, a predicted information receiving unit 102, an accident probability acquisition unit 103, and an accident probability output unit 104. These functions are realized by the hardware shown in FIG. 2 executing or processing the accident probability estimation device control program 700 according to this embodiment and various data.

[0068] The learning model storage unit 101 stores a learning model 620 that is created to calculate the probability of a distribution system accident occurring in an area within a certain period of time when prediction information of the above-mentioned events for a specified area is input based on whether or not there have been any past accidents (accident information) in each area where the distribution system is constructed, and factor information indicating the situation in each area of ​​a specified event that could be the cause of an accident.

[0069] In this embodiment, the learning model storage unit 101 is embodied as a storage device 140 .

[0070] The accident probability estimation device 100, for example, inputs learning data 610 into a learning model 620, and then adjusts the parameters that define the learning model 620 based on the difference between the result output by the learning model 620 and the input learning data 610, for example, by a method such as backpropagation, thereby training the learning model 620.

[0071] Alternatively, the accident occurrence probability estimation device 100 may perform learning of the learning model 620 so that the accident occurrence probability is output for each type of accident (ground fault accident, short circuit accident, disconnection accident, etc.).

[0072] The cause information may include information on the inspection results of the distribution system equipment installed in the area, or information indicating the period of use of the distribution system equipment installed in the area.

[0073] Furthermore, the factor information may include predetermined types of weather information such as temperature, humidity, probability of precipitation, probability of lightning, wind speed, and wind direction.

[0074] The factor information may also include information regarding geographically specific events that exist in the area, information indicating seasonally specific events that occur in the area, and information indicating the times when these events are likely to occur.

[0075] By including such information in the factor information, the accuracy of the learning model 620 is further improved, and the accident probability estimation device 100 can output a more accurate estimated cause of the accident.

[0076] Furthermore, the areas may or may not coincide with the sections defined by switches on the power distribution system, but if the areas are defined based on the sections, the accident probability output by the accident probability estimation device 100 will be in units of sections of the power distribution system, making management easier.

[0077] Alternatively, by associating areas with administrative divisions such as cities, wards, towns, villages, and settlements, it becomes possible to issue advance warnings to administrative agencies and residents of an area when the accident probability in that area output from the accident probability estimation device 100 exceeds a predetermined value.

[0078] The forecast information receiving unit 102 receives input of forecast information for the above factor information specifying an area. The forecast information receiving unit 102 acquires, for example, weather information for each day from today until one week from now, seasonal factors, regional factors, inspection results of distribution system equipment, and usage period as forecast information. Of this information, the forecast information receiving unit 102 acquires weather information from the weather information providing device 300 and acquires other information from the input device 150. Alternatively, the forecast information receiving unit 102 may acquire other information from another computer (not shown).

[0079] The accident occurrence probability acquisition unit 103 inputs the area and the above prediction information into the learning model 620, thereby acquiring the probability that an accident will occur in the distribution system in the above area within a certain period of time.

[0080] The accident occurrence probability acquisition unit 103 inputs the prediction information for each day into the learning model 620 for each day, and acquires the probability of an accident occurring for each day output from the learning model 620. When calculating the probability that an accident will occur within one week, for example, the accident occurrence probability acquisition unit 103 calculates the probability of an accident occurring within one week by calculating "1 - the product of the probability that an accident will not occur each day."

[0081] The period (fixed period) for outputting the probability of an accident from the learning model 620 may be one or more. In this manner, the accident occurrence probability for multiple periods of different lengths, such as short-term, medium-term, and long-term, can be presented to the person in charge. Alternatively, the accident occurrence probability output from the learning model 620 may be categorized by type of accident (ground fault, short-circuit fault, open circuit fault, etc.).

[0082] The accident occurrence probability output unit 104 outputs the probability of an accident occurring. The accident occurrence probability estimation device 100 outputs the accident occurrence probability by displaying, for example, a screen as shown in FIG.

[0083] In the above manner, it is possible to obtain the probability that an accident will occur in a power distribution system in a predetermined area within a certain period of time.

[0084] ==Processing flow== Next, with reference to Figures 7 and 8, we will explain the process in which the accident occurrence probability estimation device 100 of this embodiment uses learning data 610 to learn the learning model 620, and the process flow in which this learning model 620 is used to calculate the probability of an accident occurring in a distribution system.

[0085] 7 is a flowchart illustrating the processing performed by the accident occurrence probability estimation device 100 when the learning model 620 is trained using the training data 610. The timing at which the accident occurrence probability estimation device 100 executes the training of the training model 620 is not necessarily limited. For example, the accident occurrence probability estimation device 100 executes the training process of the training model 620 when new accident information is added to the cause information management table 600, when an instruction to execute training is received from a person in charge via a user interface, or when a predetermined timing occurs at predetermined intervals (such as once a month or once a year).

[0086] First, the accident occurrence probability estimation device 100 generates data that associates factor information for each area on each day in a time period with the presence or absence of an accident (accident information) in each area on each day as learning data 610 (S1000). The accident occurrence probability estimation device 100 can acquire this information from the factor information management table 600.

[0087] Next, the accident occurrence probability estimation device 100 performs a learning process for the learning model 620 using the learning data 610 (S1010).

[0088] The accident occurrence probability estimation device 100 may be configured to verify the prediction accuracy of the trained learning model 620. In this case, the accident occurrence probability estimation device 100 classifies the training data 610 into training data and verification data in advance, trains the learning model 620 using the training data, and verifies the learning model 620 using the verification data.

[0089] FIG. 8 is a flowchart illustrating the process performed by the accident occurrence probability estimation device 100 to calculate the probability that an accident will occur in a power distribution system in a predetermined area within a certain period of time.

[0090] First, the accident probability estimation device 100 acquires forecast information for each day within a certain period of time for factor information specifying an area (S2000). As described above, this forecast information may include other information such as weather information, facility inspection results, usage period, seasonal factors, and regional factors.

[0091] Next, the accident probability estimation device 100 inputs the above-mentioned prediction information for each day in the area into the learning model 620 for each day, and obtains the accident occurrence probability for each day output from the learning model 620 (S2010).

[0092] When calculating the probability that an accident will occur within one week, for example, the accident probability estimation device 100 calculates the probability that an accident will occur within one week by calculating "1 - the product of the probability that an accident will not occur each day."

[0093] Then, the accident occurrence probability estimation device 100 generates a screen showing the accident occurrence probability as shown in FIG. 9, and outputs the generated screen to the output device 160 (S2020).

[0094] In this manner, it becomes possible to determine the probability that a fault will occur in the power distribution system within a certain period of time in a specified area.

[0095] The accident occurrence probability estimation device 100, the control method and the program for the accident occurrence probability estimation device 100 according to this embodiment have been described above. According to this embodiment, it is possible to calculate the probability that an accident will occur in a power distribution system within a certain period of time in a specified area.

[0096] The above-described embodiment is intended to facilitate understanding of the present invention, and is not intended to limit the present invention. The present invention may be modified or improved without departing from the spirit thereof, and equivalents thereof are also included in the present invention. [Explanation of symbols]

[0097] 100 Accident probability estimation device 101 Learning model memory unit 102 Forecast Information Reception Department 103 Accident probability acquisition unit 104 Accident probability output section 110 CPU 120 memory 130 Communication equipment 140 Storage device 150 Input Device 160 Output Device 170 Recording medium reader 200 Power Distribution Line Automation System 300 Weather information providing device 500 Network 600 Factor Information Management Table 610 training data 620 Learning Model 621 Input Layer 622 Middle Class 623 Output Layer 630 Estimation result 700 Accident Probability Estimation Device Control Program 800 Recording Media 1000 Accident Probability Estimation System

Claims

1. An accident occurrence probability estimation device for calculating the probability of an accident occurring in a power distribution system within a certain period of time in a predetermined area, a learning model storage unit that stores a learning model created based on whether or not there have been any past accidents in the power distribution system in a plurality of areas including the predetermined area, and factor information indicating the status in each area of ​​a predetermined event that may be the cause of an accident, so as to calculate the probability that an accident will occur in the power distribution system in the area within the predetermined period when prediction information of the event for the specified area is input; a forecast information receiving unit that receives input of the forecast information specifying an area; an accident occurrence probability acquisition unit that acquires a probability of an accident occurring in the distribution system in the area within the certain period by inputting the area and the prediction information into the learning model; an accident occurrence probability output unit that outputs the probability of the accident occurring; Equipped with the fixed period includes a plurality of periods of different lengths, The accident occurrence probability estimation device, wherein the accident occurrence probability acquisition unit acquires the probability of an accident occurring for each day within the certain period from the learning model, and calculates the value obtained by subtracting the product of the probability that an accident will not occur each day from 1 for each period, thereby calculating the probability of an accident occurring for each period.

2. 2. The accident probability estimation device according to claim 1, The factor information includes a predetermined type of weather information. Accident probability estimation device.

3. 3. The accident probability estimation device according to claim 2, The weather information includes at least one of temperature, humidity, probability of precipitation, probability of lightning, wind speed, and wind direction. Accident probability estimation device.

4. The accident occurrence probability estimation device according to any one of claims 1 to 3, The accident occurrence probability estimation device, wherein the cause information includes information regarding the inspection results of the distribution system equipment installed within the area.

5. The accident occurrence probability estimation device according to any one of claims 1 to 4, The factor information includes information indicating a period of use of the distribution system equipment installed in the area. Accident probability estimation device.

6. The accident occurrence probability estimation device according to any one of claims 1 to 5, The factor information includes information on geographically specific events that exist in the area. Accident probability estimation device.

7. The accident occurrence probability estimation device according to any one of claims 1 to 6, The factor information includes information indicating seasonal events occurring in the area. Accident probability estimation device.

8. The accident occurrence probability estimation device according to any one of claims 1 to 7, The area is determined based on a section defined by switches on the power distribution system.

9. A control method for an accident occurrence probability estimation device that calculates the probability of an accident occurring in a power distribution system in a predetermined area within a certain period of time, comprising: The accident occurrence probability estimation device a first step of storing a learning model created to calculate the probability of a distribution system accident occurring in an area within a certain period when prediction information of an event that may cause an accident in an area is input, based on factor information indicating whether or not there has been a distribution system accident in the past in a plurality of areas including the specified area and the status of the event in each area that may cause an accident; a second step of receiving input of the prediction information specifying an area; a third step of acquiring a probability of an accident occurring in the distribution system in the area within the certain period by inputting the area and the prediction information into the learning model; a fourth step of outputting the probability of the accident occurring; Run the fixed period includes a plurality of periods of different lengths, In the third step, the accident occurrence probability estimation device obtains the probability of an accident occurring for each day within the certain period from the learning model, and calculates, for each period, a value obtained by subtracting the product of the probabilities that an accident will not occur for each day from 1, thereby calculating the probability of an accident occurring for each period. A method for controlling an accident probability estimation device.

10. A program for causing a computer to output the probability of an accident occurring in a power distribution system in a predetermined area within a certain period of time, a first step of storing a learning model created to calculate the probability of a distribution system accident occurring in an area within a certain period when prediction information of an event that may cause an accident in an area is input, based on factor information indicating whether or not there has been a distribution system accident in the past in a plurality of areas including the specified area and the status of the event in each area that may cause an accident; a second step of receiving input of the prediction information specifying an area; a third step of acquiring a probability of an accident occurring in the distribution system in the area within the certain period by inputting the area and the prediction information into the learning model; a fourth step of outputting the probability of the accident occurring; and causing the computer to execute the above. the fixed period includes a plurality of periods of different lengths, In the third step, the program executes a process of obtaining the probability of an accident occurring for each day within the certain period from the learning model, and calculating the value obtained by subtracting the product of the probability that an accident will not occur each day from 1 for each period, thereby calculating the probability of an accident occurring for each period.

Images