Dam temporary inspection system and dam temporary inspection program

Abstract

To enable quick and proper inspection for the presence or absence of abnormality on the earthquake occurrence around a dam.SOLUTION: A dam extra inspection system includes a seismometer 2 that monitors an earthquake around a dam facility D, a monitoring camera 3 that captures an image of the dam facility D, and a monitoring computer 5 that inspects the presence or absence of abnormality of the dam facility D on the basis of the image of the monitoring camera 3 when the seismometer 2 observes the earthquake having a predetermined seismic intensity or more. The monitoring computer 5 uses a dam inspection learning model subjected to machine learning on the basis of the past result data so that the presence or absence of abnormality of the dam facility D is output when the image of the monitoring camera 3 is input.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a dam temporary inspection system and a dam temporary inspection program for emergency / temporary inspection of the condition of a dam in the event of an earthquake. [Background technology]

[0002] Previously, if an earthquake of seismic intensity 4 or higher occurred at a location under a dam, or if a seismic intensity of 4 or higher was observed on a dam's seismometer, an emergency inspection (primary inspection) of the dam had to be carried out within one hour. This meant that employees of the power company or other entities had to come into the office to check with surveillance cameras whether there were any abnormalities, such as large cracks in the dam itself, or whether there was any abnormal water being discharged from the gates, and to check whether there were any alarms or abnormalities in the dam's quantity data on the dam monitoring control device.

[0003] Meanwhile, there is known a system that combines a camera for monitoring a dam or the like, an abnormality detection sensor for detecting abnormal situations, and a camera terminal device that performs automatic alarm processing on transmitted data when an abnormality is detected, and is connected to a central monitoring device via a digital communication line for communication (see, for example, Patent Document 1). When an abnormal situation occurs, this system controls the compression ratio of compressed data, transmission data selection, etc., and sends transmission data controlled under predetermined conditions to the central monitoring device via the digital communication line. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-333416 Summary of the Invention [Problem to be solved by the invention]

[0005] In the past, when an earthquake of seismic intensity 4 or higher occurred, companies had to check for abnormalities using surveillance cameras and dam data within one hour, and then report these inspection results to designated government agencies. This not only placed a heavy burden on employees but also risked improper inspections. In other words, checking and determining whether abnormalities exist required technical knowledge and experience, leading to variations in performance among individuals. It was also difficult to properly inspect all inspection items in the short space of one hour. Furthermore, depending on the date and time of the earthquake, it could take time to secure inspection personnel, further reducing the time available for inspections and report preparation. In contrast, in the system described in Patent Document 1, cameras monitoring dams and other structures are connected to a central monitoring device, but a person still had to review the transmitted images and check for abnormalities.

[0006] Therefore, an object of the present invention is to provide a dam emergency inspection system and a dam emergency inspection program that enable quick and appropriate inspections to determine whether or not there are any abnormalities when an earthquake occurs near a dam. [Means for solving the problem]

[0007] In order to solve the above problem, the invention of claim 1 comprises an earthquake monitoring means for monitoring earthquakes in the vicinity of a dam facility, a monitoring camera for photographing the dam facility, and when the earthquake monitoring means observes an earthquake of a predetermined seismic intensity or higher, Pre-stored sample images of abnormalities and The surveillance camera image By comparing with Inspect the dam facilities for any abnormalities and output the location, type and severity of the abnormality. An abnormality inspection means is provided. The sample image includes an image showing the abnormality location, the type of abnormality, and the degree of the abnormality. This is a dam temporary inspection system characterized by the above.

[0008] The invention of claim 2 is characterized in that, in the dam temporary inspection system described in claim 1, the abnormality inspection means uses a first dam inspection learning model that has been machine-learned based on past performance data so that when an image from the surveillance camera is input, the presence or absence of an abnormality in the dam facility is output.

[0009] The invention of claim 3 is characterized in that, in the dam temporary inspection system described in claim 1 or 2, it is provided with a report creation means for creating a specified report based on the inspection results by the abnormality inspection means.

[0010] The invention of claim 4 is characterized in that, in the dam temporary inspection system described in claims 1 to 3, it is provided with a quantity data acquisition means for acquiring dam quantity data including the dam water level of the dam facility, and the abnormality inspection means inspects the dam facility for abnormalities based on the dam quantity data.

[0011] The invention of claim 5 is characterized in that, in the dam temporary inspection system described in claim 4, the abnormality inspection means uses a second dam inspection learning model that has been machine-learned based on past performance data so that when the dam quantity data is input, the presence or absence of an abnormality in the dam facility is output.

[0012] The invention of claim 6 is a computer that monitors earthquakes around a dam facility. a receiving means for receiving the observation results from the observation means and the images from the monitoring camera that photographs the dam facility, and when the observation results received from the earthquake monitoring means are equal to or greater than a predetermined seismic intensity, Pre-stored sample images of abnormalities and Images received from the surveillance camera By comparing with Inspect the dam facilities for any abnormalities and output the location, type and severity of the abnormality. It functions as an abnormality inspection means. The sample image includes an image showing the abnormality location, the type of abnormality, and the degree of the abnormality. This is a special dam inspection program characterized by the following.

[0013] The invention of claim 7 is characterized in that, in the dam emergency inspection program described in claim 6, the abnormality inspection means uses a first dam inspection learning model that has been machine-learned based on past performance data so that when an image from the surveillance camera is input, the presence or absence of an abnormality in the dam facility is output.

[0014] The invention of claim 8 is characterized in that, in the dam emergency inspection program described in claim 6 or 7, the computer is made to function as a report creation means that creates a specified report based on the inspection results by the abnormality inspection means.

[0015] The invention of claim 9 is characterized in that, in the dam emergency inspection program described in claims 6 to 8, the receiving means receives dam quantity data including the dam water level of the dam facility, and the abnormality inspection means inspects whether or not there is an abnormality in the dam facility based on the dam quantity data.

[0016] The invention of claim 10 is characterized in that, in the dam emergency inspection program described in claim 9, the abnormality inspection means uses a second dam inspection learning model that has been machine-learned based on past performance data so that when the dam quantity data is input, the presence or absence of an abnormality in the dam facility is output. [Effects of the Invention]

[0017] According to the inventions of claims 1 and 6, when an earthquake of a predetermined seismic intensity or greater occurs near a dam facility, the dam facility is automatically inspected based on photographed images of the dam facility, making it possible to quickly inspect the facility for abnormalities when an earthquake occurs. In addition, because the dam facility is automatically inspected for abnormalities based on images without relying on human judgment, it becomes possible to inspect the facility for abnormalities appropriately and without variation when an earthquake occurs.

[0018] According to the inventions described in claims 2 and 7, the presence or absence of abnormalities in dam facilities is output using a first dam inspection learning model that has been machine-learned using images of dam facilities as an input layer, making it possible to more appropriately inspect dam facilities for abnormalities.

[0019] According to the inventions of claims 3 and 8, a predetermined report is automatically created based on the inspection results, which reduces the burden on people and enables quick reporting. In addition, because the predetermined report is created automatically without relying on people, it is possible to create reports that are free of qualitative variations and mistakes.

[0020] According to the inventions described in claims 4 and 9, the presence or absence of abnormalities in the dam facility is inspected not only based on the photographed images of the dam facility but also on dam quantity data including the dam water level, making it possible to inspect the dam facility for abnormalities more appropriately.

[0021] According to the inventions described in claims 5 and 10, the presence or absence of abnormalities in dam facilities is output using a second dam inspection learning model that has been machine-learned using dam quantity data as an input layer, making it possible to more appropriately inspect dam facilities for abnormalities. [Brief explanation of the drawings]

[0022] [Figure 1] 1 is a schematic configuration diagram showing a dam temporary inspection system according to an embodiment of the present invention. [Figure 2] FIG. 2 is a schematic block diagram of a monitoring computer of the dam temporary inspection system of FIG. 1. [Figure 3] 10A and 10B are diagrams of sample images showing a state in which a crack occurs in a dam body in an embodiment of the present invention. [Figure 4] 10A and 10B are diagrams of sample images showing a state in which abnormal water discharge occurs from the gate of the dam body in an embodiment of the present invention. [Figure 5] FIG. 3 is a functional block diagram showing a schematic configuration of a learning model for dam inspection of the monitoring computer of FIG. 2. [Figure 6] FIG. 3 is a diagram showing an example of a report sheet created by the monitoring computer of FIG. 2. DETAILED DESCRIPTION OF THE INVENTION

[0023] The present invention will be described below based on the illustrated embodiments.

[0024] Figure 1 is a schematic diagram showing a dam temporary inspection system 1 according to an embodiment of the present invention. This dam temporary inspection system 1 is a system for emergency and temporary inspection of the condition of a dam facility (dam) D in the event of an earthquake. The dam facility D includes all equipment and facilities related to dam functions, including a dam body D1, a gate (spillway gate) D2, a reservoir D3, an administration center D4, dam crest equipment, an administration road, etc. The dam facility D is also assumed to be managed and operated by an electric power company.

[0025] The dam temporary inspection system 1 includes a seismometer (earthquake monitoring means) 2, a monitoring camera 3, a water level gauge (various quantity data acquisition means) 4, and a monitoring computer (abnormality inspection means, report creation means) 5.

[0026] The seismometers 2 are measuring instruments that monitor and observe earthquakes in the vicinity of the dam facility D, and are buried or installed in the bedrock on the left and right banks of the dam main body D1, in the embankment foundation, in the inspection gallery (foundation), near the dam crest, etc. When an earthquake is observed or detected by the seismometers 2, the fact that an earthquake has been observed, along with the observed seismic intensity, acceleration, etc., are transmitted in real time to the monitoring computer 5. In this manner, in this embodiment, the seismometers 2 are installed in the dam facility D to measure the seismic intensity of the dam facility D itself, but instead of or in addition to this, an existing seismic observation station (for example, a seismic intensity observation station of the Japan Meteorological Agency) installed near the dam facility D may be used as earthquake monitoring means, and the seismic intensity, etc. from this seismic observation station may be received by the monitoring computer 5.

[0027] The monitoring cameras 3 are photographing devices for photographing the dam facilities D, and multiple cameras are installed so that they can photograph all of the monitored locations, such as the dam body D1, gate D2, reservoir D3, management office D4, dam crest equipment, and management roads. The photographed images are then sent to the monitoring computer 5 sequentially in real time.

[0028] The water level gauge 4 is a measuring instrument that measures the dam water level of the dam facility D (the water level of the reservoir D3, a river, etc.), and is installed in the reservoir D3, etc. The measured water level is then transmitted sequentially in real time to the monitoring computer 5. In this way, in this embodiment, the water level gauge 4 is installed to measure the dam water level as various dam quantity data, but in addition to this, a measuring instrument (various quantity data acquisition means) that measures and acquires various dam quantity data such as the amount of dam leakage may also be installed.

[0029] The monitoring computer 5 is a computer for monitoring the status of the dam facility D, and is installed in the management center D4. As shown in FIG. 2, the monitoring computer 5 mainly comprises an input unit 51, a display unit 52, a communication unit 53, a memory unit 54, an abnormality inspection task (abnormality inspection means) 55, a learning task 56, a report creation task (report creation means) 57, and a central processing unit 58 that controls these.

[0030] The input unit 51 is an interface for inputting various information and commands, and specifically inputs various items for the flash report (predetermined report) R described below. The display unit 52 is a display that displays various data and information, and specifically displays images of the dam facility D received from each monitoring camera 3 and the dam water level measured by the water level gauge 4. The communication unit (receiving means) 53 is an interface for communicating with the outside via the Internet network, telephone communication network, etc., and specifically communicates with the seismometer 2, monitoring camera 3, water level gauge 4, and reporting destinations described below.

[0031] The memory unit 54 mainly includes a dam database 541, dam inspection learning models (first dam inspection learning model, second dam inspection learning model) 542, and a dam inspection performance database 543. The dam database 541 is a database that stores information about the dam facility D to be monitored, and stores specification data such as the dam name, structure, area, capacity, and flood flow rate of the dam facility D, as well as details of abnormalities that occurred during past disasters, including earthquakes. The dam inspection learning models 542 and the dam inspection performance database 543 will be described later.

[0032] The abnormality inspection task 55 is a task program that inspects the dam facility D for abnormalities based on images from the monitoring camera 3 when an earthquake of a predetermined seismic intensity or greater is observed by any of the seismometers 2. In other words, when an earthquake of a predetermined seismic intensity or greater (in this embodiment, seismic intensity 4) occurs near the dam facility D, it is immediately activated, and inspects and confirms the dam facility D for abnormalities based on images of the dam facility D taken by the monitoring camera 3 (outputting the location of the abnormality, the type of abnormality, the degree of abnormality, etc.).

[0033] Specifically, the image of the dam facility D is analyzed to determine whether there are any external (visible) abnormalities, such as large-scale cracks in the dam body D1, abnormal water discharge from gate D2, etc., the collapse of the dam crest equipment, buildings, or management office D4, or large-scale collapses in the reservoir D3 or management roads, and the location, type, and severity of the abnormality are output. The presence or absence of an abnormality is determined by comparing the image of the dam facility D taken by the monitoring camera 3 with pre-stored sample images of abnormalities. For example, the presence or absence of an abnormality is determined by comparing the image with a sample image showing a state where a crack has occurred in the dam body D1, as shown in FIG. 3, or a sample image showing a state where abnormal water discharge has occurred from gate D2, as shown in FIG. 4. Here, the location of the abnormality is identified as either the dam body D1 or gate D2, the type of abnormality is identified as a crack, collapse, water leakage, etc., and the scale of the abnormality is identified as whether it is large or small, widespread or localized, etc.

[0034] The presence or absence of abnormalities in dam facility D is also checked based on dam quantity data. In other words, the presence or absence of abnormalities in dam facility D is determined not only from images of dam facility D but also from the presence or absence of sudden changes in dam quantity data, such as the dam water level measured by water level gauge 4 and the amount of dam leakage, before and after the earthquake. For example, even if it is not clear from images of dam facility D alone whether or not abnormal water is being discharged from gate D2, if the amount of dam leakage suddenly increases, it is determined that abnormal water is being discharged from gate D2, etc. Similarly, even if it is not clear whether or not a large-scale collapse has occurred in reservoir D3, if the dam water level (the water level of reservoir D3) suddenly decreases, it is determined that a large-scale collapse has occurred in reservoir D3, etc.

[0035] In this way, in this embodiment, the dam quantity data is taken into consideration to supplement and reinforce the judgment and inspection using images, but the presence or absence of an abnormality in the dam facility D may also be judged and inspected using only the dam quantity data, independently of the judgment and inspection using images. For example, if the dam water level drops sharply before and after the occurrence of an earthquake, it may be judged that a large-scale collapse has occurred in reservoir D3 or the like, or that the dam water level has dropped abnormally. Similarly, if the dam water leakage rate rises sharply before and after the occurrence of an earthquake, it may be judged that an abnormal discharge of water is occurring from gate D2 or the like, or that the dam water leakage rate has risen abnormally.

[0036] When the image from the monitoring camera 3 and the dam quantity data are input, the abnormality inspection task 55 uses a dam inspection learning model 542 that has been machine-learned based on past performance data so that the presence or absence of an abnormality in the dam facility D (the location of the abnormality, the type of abnormality, the degree of the abnormality, etc.) is output. This dam inspection learning model 542 is created by the learning task 56.

[0037] That is, the learning task 56 uses past performance data recorded and accumulated in a dam inspection performance database 543 to create a dam inspection learning model 542 using a known machine learning algorithm such as a neural network. This dam inspection performance database 543 is a database in which performance data including the results of a dam monitoring and management expert's judgment on whether or not there is an abnormality in the dam facility D based on images of the dam facility D taken by the monitoring camera 3 and dam quantity data as input information is recorded and accumulated. The past performance data includes data created based on the actual images of the dam facility D and dam quantity data and the results of judgment by the dam monitoring and management expert, as well as data created through advance training, etc.

[0038] As shown in Figure 5, this learning task 56 uses machine learning and deep learning using a neural network to create a neural network based on the performance data recorded in the dam inspection performance database 543, with the input layer including, for example, an image of the dam facility D captured by the monitoring camera 3 and various dam quantity data, the output layer including the presence or absence of an abnormality in the dam facility D (such as the location of the abnormality, the type of abnormality, and the degree of the abnormality), and the intermediate layer performing analytical processing from the input layer to the output layer.The learning task 56 then uses the performance data of the dam inspection learning model 542 as learning data to learn various parameters in the intermediate layer.In other words, the learning task 56 learns various parameters in the intermediate layer based on the image of the dam facility D and the various dam quantity data so that appropriate inspection results are determined and output.

[0039] In this embodiment, the dam inspection learning model 542 thus serves as both the first dam inspection learning model and the second dam inspection learning model in order to inspect and determine whether or not there is an abnormality in the dam facility D based on images of the dam facility D and the dam quantity data. In contrast, as described above, when the judgment and inspection of whether or not there is an abnormality based only on images of the dam facility D and the judgment and inspection of whether or not there is an abnormality based only on the dam quantity data are performed independently and separately, the first dam inspection learning model and the second dam inspection learning model are used for each judgment and inspection.

[0040] The report creation task 57 is a task program that creates a predetermined report based on the inspection results by the abnormality inspection task 55, and is started immediately when the inspection results are output from the abnormality inspection task 55. That is, in this embodiment, a flash report R that must be submitted to a predetermined reporting destination (such as a government agency) within a predetermined time (for example, one hour) after the occurrence of an earthquake with a seismic intensity of 4 or higher is created based on the inspection results by the abnormality inspection task 55.

[0041] Specifically, a sheet form for the flash report R as shown in Figure 6 is stored in advance, and based on the inspection results of the abnormality inspection task 55, a check mark is entered in either "Yes", "No", or "Unknown" in the "Damage status" column R1 of the flash report R. For example, if the inspection result output from the abnormality inspection task 55 indicates that there is an abnormality such as a large-scale crack in the dam body D1, a check mark is entered in the "Damage status" column R1 for "Whether or not there is an abnormality such as a large-scale crack in the dam body."

[0042] In addition, the dam name of the dam facility D is entered in the dam name column R2, the pre-stored department name of the electric power company is entered in the "Inspector name" column R3, the date and time when an earthquake of seismic intensity 4 or higher was observed by seismometer 2 is entered in the "Time of earthquake occurrence" column R4, the acceleration value observed by seismometer 2 is entered in the "Maximum dam acceleration" column R5, and the name of the location where the earthquake was observed (such as the name of the location where seismometer 2 is installed) and the seismic intensity observed by seismometer 2 are entered in the "Target observation point name / seismic intensity" column R6. The flash report R created in this way is then sent to the specified reporting destination.

[0043] As described above, according to this dam emergency inspection system 1, when an earthquake of a predetermined seismic intensity or greater occurs near the dam facility D, the dam facility D is inspected immediately and automatically based on the photographed image of the dam facility D, making it possible to quickly inspect whether there are any abnormalities when an earthquake occurs. In addition, because the dam facility D is inspected automatically based on the image without relying on human judgment, it becomes possible to inspect whether there are any abnormalities properly and without variation when an earthquake occurs.

[0044] Furthermore, since the presence or absence of abnormalities in the dam facility D is checked not only based on the photographed image of the dam facility D but also based on dam quantity data including the dam water level, it is possible to more accurately check the presence or absence of abnormalities in the dam facility D.

[0045] Furthermore, the dam inspection learning model 542, which has been machine-learned using images of dam facility D and dam quantity data as input layers, outputs whether or not there are any abnormalities in dam facility D, making it possible to more accurately inspect whether or not there are any abnormalities in dam facility D.

[0046] Furthermore, because Flash Reports R are automatically created and sent based on the inspection results, the burden on people can be reduced and reports can be submitted quickly. In other words, it is possible to reliably and appropriately inspect and submit Flash Reports R within a specified time after an earthquake of a specified seismic intensity or higher occurs. Furthermore, because Flash Reports R are created automatically, not by humans, it is possible to create Flash Reports R without qualitative variations or mistakes.

[0047] Although the embodiments of the present invention have been described in detail above, the specific configuration is not limited to these embodiments, and design changes within the scope of the present invention are also included in the present invention. For example, in the above embodiment, a case where a flash report R is automatically sent to a predetermined reporting destination was described, but when an earthquake of a predetermined seismic intensity or higher is observed by the seismometer 2, a pre-stored person or employee may be contacted (or a flash report R may be sent), and the flash report R may be sent to the predetermined reporting destination after a transmission command from that person is received. In this case, the report may be sent to the reporting destination designated by the person.

[0048] On the other hand, the above-described dam temporary inspection system 1 and monitoring computer 5 may be configured by installing the following dam temporary inspection program on a general-purpose computer. Specifically, the dam temporary inspection program causes the computer to function as: receiving means (communication unit 53) that receives observation results from earthquake monitoring means (seismometer 2) that monitors earthquakes in the vicinity of dam facility D, images from monitoring camera 3 that photographs dam facility D, and dam quantity data including the dam water level of dam facility D; abnormality inspection means (abnormality inspection task 55) that inspects dam facility D for abnormalities based on the images received from monitoring camera 3 and the dam quantity data when the observation results received from the earthquake monitoring means are equal to or greater than a predetermined seismic intensity; and report creation means (report creation task 57) that creates a predetermined report based on the inspection results by the abnormality inspection means. The abnormality inspection means uses a first dam inspection learning model and a second dam inspection learning model (dam inspection learning model 542) that have been machine-learned based on past performance data so that, when the images from monitoring camera 3 and the dam quantity data are input, the abnormality inspection means outputs the presence or absence of an abnormality in dam facility D. [Explanation of symbols]

[0049] 1. Dam temporary inspection system 2 Seismograph (earthquake monitoring means) 3. Surveillance cameras 4. Water level gauge (means of acquiring various data) 5. Monitoring computer (means for detecting abnormalities, means for creating reports) 53 Communication unit (receiving means) 542 Dam Inspection Learning Model (First Dam Inspection Learning Model, Second Dam Inspection Learning Model) 55 Abnormality inspection task (abnormality inspection means) 57 Report writing task (report writing means) D Dam facilities D1 Dam body D2 Gate D3 Reservoir D4 Management Office R Flash Report (prescribed report)

Claims

1. an earthquake monitoring means for monitoring earthquakes in the vicinity of the dam facility; A surveillance camera that photographs the dam facility; an abnormality inspection means for, when an earthquake of a predetermined seismic intensity or greater is observed by the earthquake monitoring means, checking for the presence or absence of abnormalities in the dam facility by comparing pre-stored sample images of abnormalities with images from the monitoring camera, and outputting the location, type and degree of the abnormality; The sample image includes an image showing the abnormality location, the type of the abnormality, and the degree of the abnormality. A dam temporary inspection system characterized by the above.

2. The abnormality inspection means uses a first dam inspection learning model that has been machine-learned based on past performance data so that, when an image from the monitoring camera is input, the presence or absence of an abnormality in the dam facility is output.

2. The dam temporary inspection system according to claim 1.

3. a report creation means for creating a predetermined report based on the inspection result by the abnormality inspection means; 3. The dam temporary inspection system according to claim 1 or 2.

4. a data acquisition means for acquiring dam data including a dam water level of the dam facility; The abnormality inspection means inspects the dam facility for abnormalities based on the dam quantity data.

4. The dam temporary inspection system according to claim 1, wherein the dam is installed in a dam.

5. The abnormality inspection means uses a second dam inspection learning model that has been machine-learned based on past performance data so that, when the dam quantity data is input, the presence or absence of an abnormality in the dam facility is output.

5. The dam temporary inspection system according to claim 4.

6. Computer, a receiving means for receiving observation results from an earthquake monitoring means for monitoring earthquakes in the vicinity of the dam facility and images from a monitoring camera for photographing the dam facility; When the observation result received from the earthquake monitoring means is equal to or greater than a predetermined seismic intensity, the means functions as an abnormality inspection means for inspecting the dam facility for abnormalities by comparing the image received from the monitoring camera with a sample image of abnormalities stored in advance, and outputting the location, type and degree of the abnormality, The sample image includes an image showing the abnormality location, the type of the abnormality, and the degree of the abnormality. A special dam inspection program.

7. The abnormality inspection means uses a first dam inspection learning model that has been machine-learned based on past performance data so that, when an image from the monitoring camera is input, the presence or absence of an abnormality in the dam facility is output.

7. The dam temporary inspection program according to claim 6.

8. Computer, functioning as a report creation means for creating a predetermined report based on the inspection results by the abnormality inspection means; 8. The dam temporary inspection program according to claim 6 or 7.

9. the receiving means receives dam quantity data including a dam water level of the dam facility, The abnormality inspection means inspects the dam facility for abnormalities based on the dam quantity data.

9. The dam temporary inspection program according to claim 6, wherein the dam temporary inspection program is a program for inspecting a dam.

10. The abnormality inspection means uses a second dam inspection learning model that has been machine-learned based on past performance data so that, when the dam quantity data is input, the presence or absence of an abnormality in the dam facility is output.

10. The dam temporary inspection program according to claim 9.

Images