Power outage investigation support system and power outage investigation support program

Abstract

To support investigation of the cause when a power outage occurs at a consumer's house.SOLUTION: A power outage investigation support system includes a smartphone 2 that takes an image of a power facility P0 installed or disposed on a utility pole P, and an investigation support computer 3 that detects appearance information including the presence or absence of an abnormality in the appearance of the power facility P0 on the basis of an image taken by the smartphone 2, and estimates the cause of the power outage, on the basis of the detected appearance information and consumer side information including the operating status of breakers H2 and H3 at a consumer's house H.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a power outage investigation support system and a power outage investigation support program for supporting an investigation into the cause of a power outage occurring at a consumer's home. [Background technology]

[0002] In the past, when a power outage occurred at a customer's home (including buildings and facilities), a repair technician would conduct an on-site survey and interview the customer to estimate and identify the cause (fault location) based on their knowledge and experience. For example, they would identify the cause as being due to a fault in the indoor equipment or a broken lead wire.

[0003] Meanwhile, a distribution line accident cause estimation device that can assist in identifying the cause of a distribution line accident is known (see, for example, Patent Document 1). This device acquires accident situation information and accident characteristic information, and estimates the cause of the distribution line accident based on the accident situation information and the accident characteristic information, and the accident situation information includes the line configuration, area characteristics, output voltage meter value, earth fault indication meter value, and zero-phase voltage value, and the accident characteristic information includes the mode and dynamics of the distribution line accident. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2022-131272 A Summary of the Invention [Problem to be solved by the invention]

[0005] Conventionally, the cause location was estimated and identified by the person in charge based on his / her knowledge and experience, so there was a risk that the accuracy varied depending on the person in charge and it took a long time. Moreover, if the knowledge and experience of the person with a wealth of knowledge and experience was not passed on to the next person, there was a risk that the cause location could not be estimated and identified. On the other hand, the device of Patent Document 1 can estimate the cause of the accident in the power distribution line itself, but when a power outage occurs at a consumer's house, it cannot estimate whether the cause is due to the consumer's house side or the power supply side.

[0006] Therefore, an object of the present invention is to provide a power outage investigation support system and a power outage investigation support program that can support an investigation into the cause when a power outage occurs at a consumer's home. [Means for solving the problem]

[0007] In order to solve the above problem, the invention of claim 1 is a power outage investigation support system for supporting investigation of the cause when a power outage occurs at a consumer's premises, characterized in that it comprises: an imaging means for photographing power equipment erected or installed on a utility pole; an appearance information detection means for detecting appearance information including the presence or absence of any abnormality in the appearance of the power equipment based on an image photographed by the imaging means; and a cause estimation means for estimating the cause of the power outage based on the appearance information detected by the appearance information detection means and consumer side information including the operating status of a breaker at the consumer's premises.

[0008] The invention of claim 2 is characterized in that, in the power outage investigation support system described in claim 1, the cause estimation means estimates the cause of the power outage based on power supply side information including voltage detection results for the power equipment.

[0009] The invention of claim 3 is characterized in that, in the power outage investigation support system described in claim 1, the cause estimation means estimates the cause of the power outage based on surrounding information including whether or not a power outage has occurred at multiple consumer homes within a specified area.

[0010] The invention of claim 4 is characterized in that, in the power outage investigation support system described in claims 1 to 3, the consumer side information includes at least one of the voltage status and insulation status of the electric wire connected to the breaker.

[0011] The invention of claim 5 is characterized in that, in the power outage investigation support system described in claim 1, the appearance information detection means uses a learning model for appearance information detection that has been machine-learned based on past performance data so that when an image captured by the photographing means is input, the appearance information is output.

[0012] The invention of claim 6 is characterized in that, in the power outage investigation support system described in claim 1, the cause estimation means uses a cause estimation learning model that has been machine-learned based on past performance data so that when the appearance information and the consumer side information are input, the cause of the power outage is output.

[0013] The invention of claim 7 is a power outage investigation support program for supporting investigation of the cause when a power outage occurs at a consumer's premises, characterized in that the computer is made to function as an appearance information detection means for detecting appearance information including the presence or absence of any abnormality in the appearance of power equipment erected or installed on a utility pole based on an image of the power equipment, and a cause estimation means for estimating the cause of the power outage based on the appearance information detected by the appearance information detection means and consumer side information including the operating status of a breaker at the consumer's premises.

[0014] The invention of claim 8 is characterized in that, in the power outage investigation support system described in claim 7, the appearance information detection means uses a learning model for appearance information detection that is machine-learned based on past performance data so that when an image of the power equipment is input, the appearance information is output.

[0015] The invention of claim 9 is characterized in that, in the power outage investigation support system described in claim 7, the cause estimation means uses a cause estimation learning model that has been machine-learned based on past performance data so that when the appearance information and the consumer side information are input, the cause of the power outage is output. Effect of the Invention

[0016] According to the inventions described in claims 1 and 7, the cause of the power outage is automatically estimated based on appearance information such as whether there is any abnormality in the power equipment on the utility pole and the operating status of the breaker at the customer's home. Therefore, it is possible to quickly and properly identify the cause without relying on human knowledge or experience by simply checking and investigating in detail only the estimated cause. Moreover, since the appearance information is automatically detected based on the image of the power equipment, it is possible to quickly and properly detect and obtain the appearance information without relying on human knowledge or experience. In this way, it is possible to support the investigation of the cause when a power outage occurs at a customer's home. Moreover, since the cause is estimated based on the appearance information of the power equipment and the operating status of the breaker, it is possible to properly estimate whether the cause is on the power supply side or the customer's home side.

[0017] According to the invention described in claim 2, the cause of the power outage is estimated based on power supply side information including the voltage detection results for the power equipment, making it possible to make an appropriate estimation according to the electrical connection status and power supply status of the power equipment.

[0018] According to the invention described in claim 3, the location of the cause of the power outage is estimated based on surrounding information such as whether or not a power outage has occurred at multiple consumer homes within a specified area, making it possible to make an appropriate estimation according to the surrounding circumstances and conditions.

[0019] According to the invention described in claim 4, the cause of the power outage is estimated based on the voltage and insulation conditions of the wires connected to the breaker, making it possible to make appropriate estimates according to a variety of events.

[0020] According to the inventions described in claims 5 and 8, appearance information is output using a machine-learned learning model for appearance information detection, making it possible to obtain more accurate appearance information and more accurately estimate the cause of the power outage.

[0021] According to the inventions described in claims 6 and 9, the cause of the power outage is output using a machine-learned learning model for cause estimation, making it possible to more accurately estimate the cause of the power outage. [Brief description of the drawings]

[0022] [Figure 1] 1 is a schematic configuration diagram showing a power outage investigation support system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a schematic configuration block diagram showing a smartphone of the power outage investigation support system of FIG. 1. [Diagram 3] 2 is a schematic configuration block diagram showing an investigation support computer of the power outage investigation support system of FIG. 1. [Figure 4] 4 is a diagram showing an example of an image to be detected in an appearance information detection task of the research support computer in FIG. 3. [Diagram 5] FIG. 4 is a functional block diagram showing a schematic configuration of a learning model for detecting appearance information of the research support computer of FIG. 3. [Figure 6] FIG. 4 is a functional block diagram showing a schematic configuration of a cause estimation learning model of the investigation support computer of FIG. 3. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] Hereinafter, the present invention will be described based on the illustrated embodiment.

[0024] 1 is a schematic diagram showing a power outage investigation support system 1 according to an embodiment of the present invention. This power outage investigation support system 1 is a system for supporting an investigation into the cause (investigation to identify the cause) when a power outage occurs at a consumer's residence H, and mainly includes a smartphone 2 and an investigation support computer 3, which are capable of communicating with each other.

[0025] The smartphone 2 is a multi-function mobile terminal carried by a person in charge M (e.g., a repair person at an electric power company) who investigates the cause of a power outage, and for example, the person in charge M carries it to the site after receiving a call from a consumer informing them of the occurrence of a power outage. This smartphone 2 has the same configuration as a normal smartphone, except for the investigation support function described below. Therefore, the configuration related to the investigation support function will be mainly described below. As shown in FIG. 2, this smartphone 2 includes a touch panel 21, a communication unit 22, a camera (imaging means) 23, an assistance application 24, and a central processing unit 25 that controls these.

[0026] The touch panel 21 is a display unit and an input unit for displaying various information and images and inputting various information and commands. For example, it displays an information input screen of the support application 24 described later, and inputs consumer side information and power supply side information described later. The communication unit 22 is an interface for communicating with the outside, and for example, it transmits power outage related information described later to the investigation support computer 3 and receives estimated cause locations described later from the investigation support computer 3.

[0027] The camera 23 is an imaging device built into or mounted on the smartphone 2, and mainly captures images of power equipment erected or disposed on the utility pole P. As shown in Fig. 4, the power equipment erected or disposed on the utility pole P includes a high-voltage line P1, an electric lighting line P2, a service line P3, a fuse P4, a pole transformer P5, a high-voltage cutout P6, a low-voltage cutout P7, a terminal box (not shown), and the like, and when these are collectively described, they will be referred to as power equipment P0 below.

[0028] The support application 24 is an application program constituting an investigation support function, which is a function for supporting investigation of the cause of a power outage. That is, the support application 24 performs the investigation support function by transmitting information and data on the power outage (power outage-related information) to the investigation support computer 3, receiving the estimated cause location from the investigation support computer 3, and displaying it on the touch panel 21.

[0029] Specifically, when started, first, an information input screen is displayed on the touch panel 21. This information input screen is a screen (input sheet screen) on which power outage-related information can be input, and the following information and data can be input along with the input sheet. First, it is an image of the power equipment P0 on the utility pole P taken by the camera 23, for example, the image shown in FIG. 4. Here, the image may be a single image showing all the power equipment P0, or may be multiple images of each piece of power equipment P0, and may be a still image or a video image.

[0030] Secondly, the consumer side information indicates information at the consumer's house H, and includes the following information. That is, the information includes the types of the breakers H2 and H3 at the consumer's house H (whether or not there is a phase loss protection function, which is a function to protect when the neutral wire and the ground wire are broken), the operation status of the breakers H2 and H3 (whether or not they have been operated), the voltage status of the electric wires and distribution lines connected to the breakers H2 and H3 (whether the secondary voltage is good, whether there is a phase loss, etc.), and the insulation status of the electric wires (results of insulation measurement, etc.). Here, the symbol H2 in FIG. 1 is a breaker with a leakage current breaker, and the symbol H3 is a breaker that only has a function to cut off the wiring. In this embodiment, the lead-in line P3 is connected to the power meter H1, and power is supplied from the power meter H1 to the load (such as a home appliance) H4 via the breaker H2 with a leakage current breaker and the breaker H3, but other connections (supply paths) may be used.

[0031] Third, there is power supply side information that indicates information related to the power equipment P0, and includes the following information: the voltage inspection result (pass / fail) for the terminal box, the voltage inspection result (pass / fail) for the load side of the fuse P4, whether the fuse P4 is blown, whether the service line P3 is broken, whether the high voltage cutout P6 or low voltage cutout P7 has been activated, etc. Other information and data can also be entered as power outage related information.

[0032] When such power outage related information is input and a transmission command (transmission button) is entered, the power outage related information is transmitted to the investigation support computer 3.

[0033] The investigation support computer 3 is a computer that estimates the location of the cause of a power outage based on the power outage-related information received from the smartphone 2, and is disposed in a management center C that manages and monitors all the power facilities P0. As shown in Fig. 3, it mainly includes an input unit 31, a display unit 32, a communication unit 33, a memory unit 34, an appearance information detection task (appearance information detection means) 35, a cause estimation task (cause estimation means) 36, a learning task 37, and a central processing unit 38 that controls these.

[0034] The input unit 31 is an interface for inputting various information and commands, and specifically, inputs a command to start the cause estimation task 36. Here, in this embodiment, as described below, a case will be described in which power outage related information received from the smartphone 2 is used in the cause estimation task 36, but the power outage related information may also be input in the input unit 31. The display unit 32 is a display that displays various data and information, and specifically, displays an image of the power facility P0 received from the smartphone 2 and displays the location of the cause estimated by the cause estimation task 36.

[0035] The communication unit 33 is an interface for communicating with the outside via the Internet network, telephone communication network, etc., and specifically receives power outage-related information from the smartphone 2 and transmits the location of the cause estimated by the cause estimation task 36 to the smartphone 2. As will be described later, the communication unit 33 receives surrounding information such as whether or not a power outage has occurred at a plurality of consumer homes H within a predetermined area (for example, within a supply range from the same utility pole P or the same pole-mounted transformer P5) from a server that manages the occurrence of power outages at consumer homes H, for example, a server of a customer center that receives information from consumers that a power outage has occurred. Such surrounding information may be transmitted from the server to the investigation support computer 3 as needed or periodically, or may be transmitted from the server in response to a transmission request from the investigation support computer 3.

[0036] The memory unit 34 mainly includes a consumer database (consumer information storage means) 341, a learning model for detecting appearance information 342, a performance database for detecting appearance information 343, a learning model for causing estimation 344, and a performance database for causing estimation 345. Here, the consumer database 341 will be explained, and the learning models 342, 344 and the performance databases 343, 345 will be explained later.

[0037] The consumer database 341 is a database that stores information about each consumer, and for each consumer's contract number, stores the contractor's name, location, contact information (telephone number, email address, etc.), the pole number of the utility pole P that supplies power to the consumer's home H, transformer information, etc. In this manner, in this embodiment, information is stored for each consumer, but, for example, information about the consumer to which power is supplied from the utility pole P may be stored for each utility pole number of the utility pole P.

[0038] The appearance information detection task 35 is a task program that detects appearance information including the presence or absence of an abnormality in the appearance of the power equipment P0 based on an image of the power equipment P0 on the utility pole P captured by the camera 23 and received from the smartphone 2. That is, when the cause estimation task 36 is started, the task program is started, and detects where and what kind of abnormality in the appearance of the power equipment P0 exists, or whether there is no abnormality at all, based on an image of the power equipment P0 included in the power outage-related information that is an input parameter at the time of starting the cause estimation task 36. Here, an abnormality is a phenomenon or event that is not usually (until now) observed, and also includes bird droppings that are simply attached but do not cause a problem in the power supply by the power equipment P0 (does not cause a power outage).

[0039] Specifically, in this embodiment, an image of the power equipment P0 in a normal state (when there is no abnormality) is stored as a normal image, and appearance information such as where and what kind of appearance abnormality exists is detected by comparing this normal image with an image (current image) received from the smartphone 2. For example, appearance information such as a break in the service line P3, a blown fuse P4, damage to the high voltage cutout P6 or low voltage cutout P7, or operation of the high voltage cutout P6 or low voltage cutout P7 is detected.

[0040] In this appearance information detection task 35, a learning model 342 for detecting appearance information that has been machine-learned based on past performance data is used so that appearance information is output when an image (current image) of the power equipment P0 is input. This learning model 342 for detecting appearance information is created by the learning task 37.

[0041] That is, the learning task 37 uses past performance data recorded and accumulated in a performance database 343 for detecting appearance information to create a learning model 342 for detecting appearance information by a known machine learning algorithm such as a neural network. This performance database 343 for detecting appearance information is a database in which performance data including an image of the power equipment P0 as input information and appearance information confirmed by an expert or knowledgeable person in the management and inspection of the power equipment P0 (such as the presence or absence of an abnormality in the appearance of the power equipment P0) is recorded and accumulated. The past performance data includes an image of the actual power equipment P0, the appearance information actually confirmed, and data created in advance training, etc.

[0042] As shown in Fig. 5, this learning task 37 uses machine learning and deep learning using a neural network to create a neural network with, for example, an image of the power equipment P0 as an input layer, appearance information as an output layer, and analysis processing from the input layer to the output layer as an intermediate layer based on the performance data recorded in the performance database 343 for detecting appearance information. Then, the learning task 37 uses the performance data of the learning model 342 for detecting appearance information as learning data to learn various parameters in the intermediate layer. That is, the learning task 37 learns various parameters in the intermediate layer so that appearance information such as where and what kind of appearance abnormality is present in the power equipment P0 is appropriately output based on the image of the power equipment P0.

[0043] The cause estimation task 36 is a task program that estimates the cause of the power outage (outputs the estimated cause) based on the appearance information detected by the appearance information detection task 35 and the consumer side information, etc. That is, when the power outage related information is received from the smartphone 2, the cause estimation task 36 is started, and first, the appearance information detection task 35 is started, and as described above, the appearance information including the presence or absence of abnormality in the appearance of the power equipment P0 is detected based on the image of the power equipment P0 in the power outage related information. Next, the cause of the power outage is estimated based on the detected appearance information, the consumer side information and the power supply side information in the power outage related information, and the surrounding information received from the server of the customer center, etc. Specifically, the cause of the power outage is estimated as follows. If the consumer-side information indicates that breakers H2 and H3 have been activated and that breakers H2 and H3 have open-phase protection functions, it is presumed that the cause of the problem may be located both in the power equipment P0 (power supply side) and in the consumer's home H (consumer side). If the consumer side information includes that the secondary voltage measurement results of breakers H2 and H3 are good, but the insulation measurement results are poor, it is presumed that the cause of the problem is in consumer home H. If the customer-side information includes that breakers H2 and H3 are not operating and the voltage measurement results of breakers H2 and H3 show one phase is missing (faulty), the cause is presumed to be a broken wire in service line P3 or a blown fuse P4. ·If the power supplier information shows that the voltage test result for the terminal box is good, but the voltage test result for the load side of fuse P4 is bad, it is assumed that the cause is a blown fuse P4. If the customer-side information indicates that breakers H2 and H3 are not operating and that voltage measurements of breakers H2 and H3 indicate a complete loss of phases (fault), it is assumed that the cause is a broken line P3, a damaged power meter H1, or a blown high-voltage fuse. If the surrounding information includes information that a power outage has occurred at multiple consumer homes H within the supply area of ​​the same pole transformer P5, it is presumed that the cause is a blown high-voltage fuse. If the customer-side information includes the results of voltage measurements on breakers H2 and H3 being poor, and the external information includes information that there is a break in service line P3, it is presumed that the cause is a break in service line P3. · If the customer-side information includes the results of voltage measurements on breakers H2 and H3 being bad, and the appearance information includes information that fuse P4 is not blown and that there is discoloration inside the terminal box, then it is presumed that the cause is poor contact in the terminal box. - If the customer-side information includes the results of voltage measurements on breakers H2 and H3 being poor, the surrounding information includes the occurrence of power outages at multiple customer homes H within the supply area from the same pole transformer P5, and the external appearance information includes the presence of damage to cutouts P6 and P7, then it is presumed that the cause is damage to cutouts P6 and P7. If the customer-side information includes information that the voltage measurement results of breakers H2 and H3 are good, that breakers H2 and H3 have open-phase protection functions, and that breakers H2 and H3 sometimes operate, it is assumed that the cause is poor contact of the ground wire. - If the customer-side information includes the results of the voltage measurements of breakers H2 and H3 being poor, the surrounding information includes the occurrence of power outages at multiple customer homes H within the supply area from the same pole transformer P5, and the external appearance information includes the fact that the high-voltage fuse is not operational, then it is presumed that the cause is a break in the high-voltage downline. If the customer-side information includes poor voltage measurement results for breakers H2 and H3, and the surrounding information includes information that a power outage has occurred at multiple customer homes H within the supply area from the same pole transformer P5, and that there is also a power outage in another bank (supply area from another pole transformer P5), then it is presumed that the cause is a break in the high-voltage line P1.

[0044] Such a cause estimation task 36 uses a cause estimation learning model 344 that is machine-learned based on past performance data so that the cause location of the power outage (estimated cause location) is output when appearance information, consumer side information, power supplier side information, and surrounding information are input. This cause estimation learning model 344 is created by a learning task 37.

[0045] That is, the learning task 37 uses past performance data recorded and accumulated in a cause estimation performance database 345 to create a cause estimation learning model 344 by a known machine learning algorithm such as a neural network. This cause estimation performance database 345 is a database in which performance data including appearance information, consumer side information, power supply side information, and surrounding information as input information, and the cause of the power outage confirmed and identified by experts and others who are skilled in power outage cause investigation, is recorded and accumulated. The past performance data includes the actual appearance information, consumer side information, power supply side information, and surrounding information, the actually confirmed cause of the power outage, as well as data created in advance training, etc.

[0046] As shown in Fig. 6, this learning task 37 uses machine learning and deep learning using a neural network to create a neural network based on the performance data recorded in the cause estimation performance database 345, with, for example, appearance information, consumer side information, power supply side information, and surrounding information as an input layer, the cause of the power outage as an output layer, and analysis processing from the input layer to the output layer as an intermediate layer. Then, the learning task 37 uses the performance data of the cause estimation learning model 344 as learning data to learn various parameters in the intermediate layer. That is, the learning task 37 learns various parameters in the intermediate layer based on the appearance information, consumer side information, power supply side information, and surrounding information so that the cause of the power outage is appropriately output.

[0047] In this way, the location of the cause of the power outage is estimated and output, and this estimated location of the cause is transmitted to the smartphone 2.

[0048] Next, the action and operation of the power outage investigation support system 1 configured as above and a power outage investigation support method using the power outage investigation support system 1 will be described.

[0049] First, upon receiving a notification from a consumer that a power outage has occurred, a person in charge M of the power company goes to the site with a smartphone 2. Then, the person in charge M takes a picture of the power equipment P0 on the utility pole P that supplies power to the consumer's home H that has experienced a power outage, using the camera 23 of the smartphone 2. The person in charge M also obtains consumer-side information by inspecting the breakers H2 and H3 of the consumer's home H, and obtains power supply-side information by inspecting the power equipment P0. Next, the person in charge M starts the support app 24 on the smartphone 2, inputs information and data about the power outage (power outage-related information) according to the information input screen displayed on the smartphone 2, and transmits it to the investigation support computer 3.

[0050] In response to this, the cause estimation task 36 is started in the investigation support computer 3, and the location of the cause of the power outage is estimated (the estimated cause location is output) as described above. Then, the estimated cause location is transmitted to the smartphone 2, and the estimated cause location is displayed on the touch panel 21 of the smartphone 2.

[0051] As described above, according to the power outage investigation support system 1 and the power outage investigation support method, the cause of the power outage is automatically estimated based on the appearance information, such as whether there is any abnormality in the appearance of the power equipment P0 on the utility pole P, and the operating status of the breakers H2 and H3 of the consumer's home H. Therefore, it is possible to quickly and properly identify the cause without relying on human knowledge or experience, by simply checking and investigating in detail only the estimated cause. Moreover, since the appearance information is automatically detected based on the image of the power equipment P0, it is possible to quickly and properly detect and obtain the appearance information without relying on human knowledge or experience. In this way, it is possible to support the investigation of the cause when a power outage occurs at the consumer's home H. Moreover, since the cause is estimated based on the appearance information of the power equipment P0 and the operating status of the breakers H2 and H3, it is possible to properly estimate whether the cause is on the power supply side or the consumer's home H side.

[0052] In addition, since the cause of the power outage is estimated based on power supply side information such as the result of voltage detection for the power equipment P0, it is possible to make an appropriate estimation according to the electrical connection status and power supply status of the power equipment P0. Furthermore, since the cause of the power outage is estimated based on surrounding information such as whether or not a power outage has occurred in multiple consumer homes H within a specified area, it is possible to make an appropriate estimation according to the surrounding situation and state. Moreover, since the cause of the power outage is estimated based on the voltage status and insulation status of the electric wires connected to the breakers H2 and H3, it is possible to make an appropriate estimation according to various events.

[0053] In addition, since the appearance information is output using the machine-learned learning model for appearance information detection 342, it is possible to acquire more appropriate appearance information and more appropriately estimate the cause of the power outage. Similarly, since the cause of the power outage is output using the machine-learned learning model for cause estimation 344, it is possible to more appropriately estimate the cause of the power outage.

[0054] Although the embodiment of the present invention has been described above in detail, the specific configuration is not limited to this embodiment, and even if there are design changes within the scope of the present invention, they are included in the present invention. For example, in the above embodiment, the investigation support computer 3 is provided with the cause estimation task 36, and the investigation support computer 3 estimates the cause of the power outage. However, the smartphone 2 may be provided with the cause estimation task 36, and the smartphone 2 may estimate the cause of the power outage.

[0055] On the other hand, the above-described power outage investigation support system 1 and investigation support computer 3 may be configured by installing the following power outage investigation support program on a general-purpose computer.

[0056] That is, the power outage investigation support program is for supporting investigation of the cause when a power outage occurs at a consumer's home H, and causes a computer to function as: appearance information detection means (appearance information detection task 35) for detecting appearance information including the presence or absence of an abnormality in the appearance of power equipment P0 based on an image taken of the power equipment P0 erected or disposed on a utility pole P; and cause estimation means (cause estimation task 36) for estimating the cause of the power outage based on the appearance information detected by the appearance information detection means, consumer side information including the operating status of the breaker at the consumer's home H, power supply side information including the voltage detection result for the power equipment P0, and surrounding information including whether a power outage has occurred at a plurality of consumer homes within a specified area. The appearance information detection means uses a learning model 342 for appearance information detection machine-learned based on past performance data so that appearance information is output when an image taken of the power equipment P0 is input; and the cause estimation means uses a learning model 344 for cause estimation machine-learned based on past performance data so that the cause of the power outage is output when the appearance information, consumer side information, power supply side information, and surrounding information are input. [Explanation of symbols]

[0057] 1. Power outage investigation support system 2. Smartphone 23 Camera (photography means) 3. Research Support Computer 342 Learning Model for Appearance Information Detection 343 Appearance Information Detection Performance Database 344 Learning Model for Cause Estimation 345 Database of results for cause estimation 35 Appearance information detection task (appearance information detection means) 36 Cause inference task (cause inference means) M person in charge P utility pole P0 Power equipment H Consumer's house H1 Power Meter H2 Earth leakage circuit breaker H3 Breaker H4 Load (household appliances, etc.)

Claims

1. A power outage investigation support system for supporting investigation of the cause of a power outage occurring at a consumer's home, An imaging means for imaging an electric power facility installed or disposed on a utility pole; an appearance information detection means for detecting appearance information including the presence or absence of an abnormality in the appearance of the power facility based on the image captured by the imaging means; A cause estimation means for estimating a cause of the power outage based on the appearance information detected by the appearance information detection means and consumer side information including an operating status of a breaker in the consumer's home; A power outage investigation support system comprising:

2. The cause estimation means estimates a cause of the power outage based on power supply side information including a voltage detection result for the power equipment. The power outage investigation support system according to claim 1 .

3. The cause estimation means estimates a cause of the power outage based on surrounding information including whether a power outage has occurred in a plurality of consumer homes within a predetermined area. The power outage investigation support system according to claim 1 .

4. The customer side information includes at least one of a voltage state and an insulation state of an electric wire connected to the breaker. The power outage investigation support system according to any one of claims 1 to 3.

5. the appearance information detection means uses a learning model for detecting appearance information that is machine-learned based on past performance data so that the appearance information is output when an image captured by the imaging means is input; The power outage investigation support system according to claim 1 .

6. The cause estimation means uses a learning model for cause estimation that is machine-learned based on past performance data so that the cause location of the power outage is output when the appearance information and the consumer side information are input. The power outage investigation support system according to claim 1 .

7. A power outage investigation support program for supporting an investigation into the cause of a power outage occurring at a consumer's home, comprising: An appearance information detection means for detecting appearance information including the presence or absence of abnormality in the appearance of electric power equipment installed or disposed on a utility pole based on an image of the electric power equipment; a cause estimation means for estimating a cause of the power outage based on the appearance information detected by the appearance information detection means and consumer side information including an operating status of a breaker in the consumer's home; A power outage investigation support program characterized by functioning as a power outage investigation support program.

8. The appearance information detection means uses a learning model for detecting appearance information that has been machine-trained based on past performance data so that the appearance information is output when an image of the power equipment is input.

8. The power outage investigation support program according to claim 7.

9. The cause estimation means uses a learning model for cause estimation that is machine-learned based on past performance data so that the cause location of the power outage is output when the appearance information and the consumer side information are input.

8. The power outage investigation support program according to claim 7.

Images