Method, device, equipment and storage medium for monitoring hydropower station equipment

Abstract

The present application is applicable to the field of data processing technology, and provides a method, apparatus, equipment and storage medium for monitoring hydropower station equipment, including: using ghost imaging technology to collect light field distribution information and reflected light intensity information of equipment monitoring points; inputting the light field distribution information and reflected light intensity information into a trained leakage monitoring model for processing to obtain monitoring results of the equipment monitoring points, and the leakage monitoring model is obtained by training the initial model using a sample training set. In the above scheme, ghost imaging technology is used to collect light field distribution information and reflected light intensity information of equipment monitoring points, so that the collected light field distribution information and reflected light intensity information of the equipment monitoring points are not affected by the on-site light intensity, and accurately reflect the current status of the equipment monitoring points. On this basis, by processing the light field distribution information and reflected light intensity information based on the leakage monitoring model, it is possible to accurately identify whether the hydropower station equipment is leaking, thereby improving the accuracy of the monitoring results.

Description

Technical Field

[0001] The present application belongs to the field of data processing technology, and in particular relates to a method, apparatus, equipment and storage medium for monitoring hydropower station equipment. Background Art

[0002] Hydropower stations generally refer to hydroelectric plants. As clean electricity producers, they place extremely high demands on safety within their production areas. Within these areas, hydropower stations often feature generator sets, gas-insulated metal-enclosed switchgear (GIS), transformers, and other equipment containing oil, water, and gas, as well as various pipelines. These devices, exposed to wind, water, electricity, and magnetism, are prone to leakage over long periods of operation, posing a significant threat to production operations.

[0003] Prior art uses deep convolutional neural networks to identify images of hydropower station equipment to monitor for leaks. However, because images of hydropower station equipment are easily affected by on-site light intensity, deep convolutional neural networks cannot accurately determine whether the equipment is leaking, resulting in inaccurate monitoring results. Summary of the Invention

[0004] In view of this, embodiments of the present application provide a method, apparatus, device, and storage medium for monitoring hydropower station equipment to solve the problem of inaccurate monitoring of whether hydropower station equipment is leaking in the prior art.

[0005] A first aspect of an embodiment of the present application provides a method for monitoring hydropower station equipment, the method comprising:

[0006] Ghost imaging technology is used to collect light field distribution information and reflected light intensity information of the equipment monitoring point; the light field distribution information and reflected light intensity information are input into a trained leakage monitoring model for processing to obtain the monitoring results of the equipment monitoring point. The leakage monitoring model is obtained by training the initial model using the sample training set.

[0007] In this solution, ghost imaging technology is used to collect light field distribution and reflected light intensity information at equipment monitoring points. This information is unaffected by the intensity of the light at the site, accurately reflecting the current status of the equipment monitoring points. Furthermore, by processing this information based on a leakage monitoring model, it can accurately identify whether hydropower station equipment is leaking, thereby improving the accuracy of monitoring results.

[0008] Optionally, the light field distribution information and the reflected light intensity information are input into a trained leakage monitoring model for processing to obtain monitoring results of the equipment monitoring point, including: extracting the structured light features of the light field distribution information through the leakage monitoring model; extracting the intensity features of the reflected light intensity information through the leakage monitoring model; determining the contour features of the equipment monitoring point based on the structured light features and the intensity features; and determining the monitoring results based on the contour features.

[0009] Optionally, the monitoring result includes that the equipment monitoring point is in a normal state or that the equipment monitoring point is in an abnormal state. The light field distribution information and the reflected light intensity information are input into a trained leakage monitoring model for processing to obtain the monitoring result of the equipment monitoring point. The method also includes: when the monitoring result is that the equipment monitoring point is in an abnormal state, generating a ghost imaging image corresponding to the equipment monitoring point according to the light field distribution information and the reflected light intensity information; comparing the ghost imaging image with a preset leakage image; and determining the final monitoring result of the monitored part based on the comparison result.

[0010] Optionally, the method also includes: obtaining a sample training set, the sample training set including multiple sample light field distribution information, sample reflected light intensity information corresponding to each sample light field distribution information, and sample monitoring results corresponding to each sample light field distribution information; inputting the sample light field distribution information and the sample reflected light intensity information in the sample training set into the initial model for processing to obtain the actual monitoring results of the sample light field distribution information; calculating the loss value between the actual monitoring results and the sample monitoring results corresponding to the sample light field distribution information according to a preset loss function; when it is detected that the loss value is greater than a preset threshold, adjusting the model parameters of the initial model, and continuing to train the initial model using the sample training set; when it is detected that the loss value is less than or equal to the preset threshold, stopping training the initial model, and determining the trained initial model as the leakage monitoring model.

[0011] Optionally, ghost imaging technology is used to collect light field distribution information and reflected light intensity information of the equipment monitoring point, including: obtaining the initial monitoring result of the leakage sensor; when the initial monitoring result is that the equipment monitoring point is abnormal, the ghost imaging method is used to collect light field distribution information and reflected light intensity information.

[0012] Optionally, the final monitoring result includes leakage at the equipment monitoring point or no leakage at the equipment monitoring point. After determining the final monitoring result of the monitored part according to the comparison result, the method further includes: when the final monitoring result is leakage at the equipment monitoring point, issuing an early warning reminder.

[0013] Optionally, when the final monitoring result is that the equipment monitoring point is leaking, an early warning reminder is issued, including: when the final monitoring result is that the equipment monitoring point is leaking, determining the leakage level of the equipment monitoring point based on the ghost imaging image; determining the early warning method based on the leakage level, and issuing an early warning based on the early warning method.

[0014] A second aspect of an embodiment of the present application provides a device for monitoring hydropower station equipment, comprising:

[0015] An acquisition unit, configured to acquire light field distribution information and reflected light intensity information of a monitoring point of the device using ghost imaging technology;

[0016] The processing unit is used to input the light field distribution information and the reflected light intensity information into the trained leakage monitoring model for processing to obtain the monitoring results of the equipment monitoring point. The leakage monitoring model is obtained by training the initial model using the sample training set.

[0017] A third aspect of an embodiment of the present application provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for monitoring hydropower station equipment as described in the first aspect above are implemented.

[0018] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for monitoring hydropower station equipment as described in the first aspect above.

[0019] A fifth aspect of the embodiments of the present application provides a computer program product, which, when executed on a terminal device, enables the terminal device to execute the steps of the method for monitoring hydropower station equipment described in the first aspect above. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0021] Figure 1 is a schematic flow chart of a method for monitoring hydropower station equipment provided by an exemplary embodiment of the present application;

[0022] Figure 2 This is a schematic diagram of the ghost imaging principle provided by this application;

[0023] Figure 3 is a specific flow chart of step S102 of a method for monitoring hydropower station equipment shown in another exemplary embodiment of the present application;

[0024] Figure 4is a schematic flow chart of a method for monitoring hydropower station equipment shown in yet another exemplary embodiment of the present application;

[0025] Figure 5 is a schematic flow chart of a method for training a leakage monitoring model provided by the present application;

[0026] Figure 6 This is a schematic diagram of a device for monitoring hydropower station equipment provided by an embodiment of the present application;

[0027] Figure 7 This is a schematic diagram of a terminal device provided in another embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0029] Hydropower stations generally refer to hydroelectric plants. As clean electricity producers, they place extremely high demands on safety within their production areas. Within these areas, hydropower stations often feature generators, GIS, transformers, and other equipment containing oil, water, and gas, as well as various pipelines. These devices, exposed to wind, water, electricity, and magnetism, are prone to leakage over long periods of operation, posing a significant threat to production operations.

[0030] For example, pipes and valves in the volute layer, water supply equipment layer, tailwater pipe layer, etc. in hydropower station equipment often leak. Only by discovering and solving this phenomenon in time can pollution be avoided and the safe operation of hydropower station equipment be guaranteed.

[0031] Prior art uses deep convolutional neural networks to identify images of hydropower station equipment to monitor for leaks. However, because images of hydropower station equipment are easily affected by on-site light intensity, they can be noisy and of poor quality in dim environments. When processed using deep convolutional neural networks, these images cannot accurately determine whether the equipment is leaking, resulting in inaccurate monitoring results.

[0032] In view of this, an embodiment of the present application provides a method for monitoring hydropower station equipment, which uses ghost imaging technology to collect light field distribution information and reflected light intensity information of the equipment monitoring points; the light field distribution information and reflected light intensity information are input into a trained leakage monitoring model for processing to obtain the monitoring results of the equipment monitoring points. The leakage monitoring model is obtained by training the initial model using a sample training set.

[0033] This solution uses ghost imaging technology to collect light field distribution and reflected light intensity information at equipment monitoring points. This information is unaffected by the intensity of the light at the site, accurately reflecting the current status of the equipment monitoring point. Furthermore, this information is processed based on a leakage monitoring model to accurately identify leaks in hydropower station equipment, thereby improving the accuracy of monitoring results.

[0034] See Figure 1 , Figure 1 This is a schematic flow chart of a method for monitoring hydropower station equipment provided by an exemplary embodiment of the present application. The execution subject of the method for monitoring hydropower station equipment provided by the present application is a device, wherein the device includes but is not limited to a vehicle-mounted computer, a tablet computer, a computer, a personal digital assistant (PDA), and other devices, and may also include various types of servers. For example, the server can be an independent server, or it can be a cloud service that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0035] like Figure 1 The method for monitoring hydropower station equipment shown may include: S101 to S102, specifically as follows:

[0036] S101: Using ghost imaging technology to collect light field distribution information and reflected light intensity information of the device monitoring point.

[0037] Equipment monitoring points refer to areas of hydropower station equipment that need to be monitored for leaks. Specifically, these can include generator sets, GIS, transformers, and other components. They can also include areas containing pipes and valves, such as the volute, water supply, and tailwater pipes.

[0038] Ghost imaging, also known as correlation imaging or two-photon imaging, is a novel imaging technique that uses two-photon compound detection to recover the spatial information of an object under test (such as a device monitoring point).

[0039] For easier understanding, see Figure 2 , Figure 2 This is a schematic diagram of the ghost imaging principle provided by this application.

[0040] like Figure 2As shown, the light source is used to generate light, the beam splitter is used to split a beam of light into two beams of light, the object is the equipment monitoring point, the bucket detector is used to detect the signal light after it hits the object (i.e., the equipment monitoring point), and the charge coupled device (CCD) camera is used to detect the idle light after it hits the object (i.e., the equipment monitoring point). Figure 2 The symbol in the lower right corner indicates that the detected signal light and idle light are correlated, and the image of the object (i.e., the equipment monitoring point) can be restored after the correlation operation.

[0041] The light source may be infrared light, laser light, array light, etc., without limitation. The signal light includes reflected light intensity information, and the idle light includes light field distribution information.

[0042] For example, a preset light modulator is triggered, causing it to control the light source to emit a beam of light. This beam of light is then split into two beams by a beam splitter prism when it strikes the device's monitoring point. This beam of light, called the signal light path (also known as the object arm), is then transmitted through a lens and collected by a barrel detector positioned behind the lens. Specifically, the barrel detector collects information about the intensity of the reflected light generated by the beam of light striking the device's monitoring point.

[0043] The other beam of light split by the beam splitter prism is irradiated on the CCD camera. This beam of light can be called the reference light path (also called the reference arm). The CCD camera has spatial resolution capability and scans the cross section of this light to obtain light field distribution information.

[0044] The light field distribution information includes the position and direction information of each object at the device monitoring point, and the reflected light intensity information includes the total light intensity reflected by each object at the device monitoring point.

[0045] It's worth noting that a single measurement can capture both reflected light intensity information and light field distribution information. Multiple measurements can capture multiple pieces of reflected light intensity information and light field distribution information. Correlating these multiple pieces of reflected light intensity information and light field distribution information can yield a ghost image corresponding to the device's monitoring point. In this embodiment, capturing both the light field distribution information and the reflected light intensity information at the device's monitoring point is sufficient.

[0046] In this embodiment, ghost imaging technology is directly used to collect light field distribution information and reflected light intensity information of the equipment monitoring point. Optionally, in some possible implementation methods of the present application, in order to reduce the frequency of collecting light field distribution information and reflected light intensity information of the equipment monitoring point, thereby reducing memory and saving resources, a leakage sensor can be pre-installed in the hydropower station equipment, and the equipment monitoring point can be preliminarily monitored by the leakage sensor. According to the initial monitoring results of the leakage sensor, it is determined whether it is necessary to use ghost imaging technology to collect light field distribution information and reflected light intensity information of the equipment monitoring point.

[0047] Specifically, the above S101 may include S1011 to S1012, which are as follows:

[0048] S1011: Obtaining initial monitoring results of the liquid leakage sensor.

[0049] The initial monitoring results include whether the equipment monitoring point is normal or abnormal.

[0050] For example, a liquid leakage sensor is pre-installed in the hydropower station equipment to initially monitor whether an equipment monitoring point is abnormal. For example, if the liquid leakage sensor detects water and / or oil leakage at the equipment monitoring point, the equipment monitoring point is determined to be abnormal. For another example, if the liquid leakage sensor detects no water or oil leakage at the equipment monitoring point, the equipment monitoring point is determined to be normal.

[0051] The specific process of the liquid leakage sensor monitoring the equipment monitoring point can be referred to the existing technology and will not be described in detail here.

[0052] The leak sensor transmits its initial monitoring results to the device. After receiving these initial monitoring results, the device performs different processing based on the different initial monitoring results. If the initial monitoring result indicates that the device's monitoring point is normal, the device continues to obtain the initial monitoring results of the leak sensor. If the initial monitoring result indicates that the device's monitoring point is abnormal, the device executes S1012.

[0053] S1012: When the initial monitoring result indicates that the device monitoring point is abnormal, light field distribution information and reflected light intensity information are collected using a ghost imaging method.

[0054] Exemplarily, when the initial monitoring result is determined to be an abnormality at the equipment monitoring point, the ghost imaging method is used to collect light field distribution information and reflected light intensity information. Specifically, the device triggers a preset light modulator so that the light modulator controls the light source to emit a beam of light. When this beam of light is irradiated on the dichroic prism, it is split into two beams of light by the dichroic prism. Among them, the beam of light separated by the dichroic prism is irradiated on the equipment monitoring point, and the bucket detector collects the reflected light intensity information generated after the beam of light is irradiated on the equipment monitoring point. The other beam of light separated by the dichroic prism is irradiated on the CCD camera, and the CCD camera scans the cross section of the beam of light to obtain light field distribution information.

[0055] In this implementation, a conventional liquid leakage sensor is first used to perform preliminary monitoring of the device monitoring point. Only when the initial monitoring result indicates an abnormality at the device monitoring point is ghost imaging technology used to collect relevant information. Subsequently, the collected light field distribution information and reflected light intensity information are processed using a liquid leakage monitoring model to further determine the status of the device monitoring point. This implementation, on the one hand, eliminates the need to constantly collect light field distribution and reflected light intensity information at the device monitoring point, reducing memory usage, conserving resources, and extending the lifespan of each device. On the other hand, further determining the status of the device monitoring point based on the initial monitoring results of the liquid leakage sensor greatly improves the accuracy of the monitoring results.

[0056] S102: Input the light field distribution information and the reflected light intensity information into the trained leakage monitoring model for processing to obtain the monitoring results of the equipment monitoring point.

[0057] Exemplarily, the leakage monitoring model is obtained by training an initial model using a sample training set. The sample training set includes multiple sample light field distribution information, sample reflected light intensity information corresponding to each sample light field distribution information, and sample monitoring results corresponding to each sample light field distribution information. The monitoring results include whether the device monitoring point is in a normal state or an abnormal state.

[0058] In this embodiment, the leakage monitoring model can be pre-trained by this device, or it can be pre-trained by another device and then the corresponding file for the leakage monitoring model can be ported to this device. In other words, the entity that trains the leakage monitoring model and the entity that uses it can be the same or different. For example, when another device is used to train the leakage monitoring model, after the other device completes training, it fixes the model parameters, obtains the file corresponding to the trained leakage monitoring model, and then ports this file to this device.

[0059] For example, if only one set of light field distribution information and reflected light intensity information is collected, the set of light field distribution information and reflected light intensity information is input into the trained leakage monitoring model for processing to obtain the monitoring results of the equipment monitoring point.

[0060] It is understood that the more light field distribution information and reflected light intensity information collected, the more accurate the monitoring results obtained after processing by the leakage monitoring model. For example, if multiple sets of light field distribution information and reflected light intensity information are collected, each set of light field distribution information and reflected light intensity information is input into the trained leakage monitoring model for processing to obtain monitoring results for the device monitoring point.

[0061] In this embodiment, ghost imaging technology is used to collect light field distribution and reflected light intensity information at equipment monitoring points. Even in indoor environments, this information is unaffected by the intensity of the surrounding light. Furthermore, regardless of the acquisition angle or location, comprehensive and accurate information about the equipment monitoring points can be obtained using this information, accurately reflecting the current status of the equipment monitoring points. Furthermore, by processing this light field distribution and reflected light intensity information based on a leakage monitoring model, accurate monitoring results can be obtained, enabling precise determination of leakage in hydropower station equipment and improving the accuracy of monitoring of equipment monitoring points.

[0062] See Figure 3 , Figure 3 1 is a specific flow chart of step S102 of a method for monitoring hydropower station equipment shown in another exemplary embodiment of the present application; optionally, in some possible implementations of the present application, the above S102 may include S1021 to S1024, as follows:

[0063] S1021: Extracting structured light features of light field distribution information through a liquid leakage monitoring model.

[0064] Exemplarily, the collected light field distribution information is represented in matrix form, i.e., in the form of an array. The leakage monitoring model includes an encoder, which is composed of multiple convolutional layers. For example, the encoder in this embodiment is composed of three convolutional layers.

[0065] The convolution layer performs convolution processing on the light field distribution information to extract the structured light features in the light field distribution information. The structured light features are used to represent the characteristics of each object in the device monitoring point after being illuminated by light.

[0066] S1022: Extracting intensity features of reflected light intensity information through a liquid leakage monitoring model.

[0067] For example, the collected reflected light intensity information is represented in numerical form. The intensity feature represents the intensity value generated by each object at the device monitoring point when illuminated by light. Specifically, the intensity value in the reflected light intensity information can be extracted using a leakage monitoring model. This is merely an example and is not intended to be limiting.

[0068] S1023: Determine the contour features of the equipment monitoring point based on the structured light features and the intensity features.

[0069] For example, the extracted structured light features and intensity features are correlated to obtain the contour features of the equipment monitoring point. This contour feature is used to map the outlines of various objects in the equipment monitoring point. For example, the outlines of pipes and valves in the equipment monitoring point can be mapped. If there is a leak at the equipment monitoring point, the outline of the liquid (such as oil or water) will also be mapped.

[0070] It's worth noting that if only one set of light field distribution information and reflected light intensity information is collected, the leakage monitoring model extracts the structured light features from the light field distribution information and the intensity features from the reflected light intensity information. The extracted structured light features and intensity features are then correlated to obtain the contour features of the device monitoring point.

[0071] If multiple sets of light field distribution information and reflected light intensity information are collected, such as 5 sets of light field distribution information and reflected light intensity information, the leakage monitoring model is used to extract the structured light features in each set of light field distribution information and the intensity features in each set of reflected light intensity information.

[0072] For each set of light field distribution information and reflected light intensity information, the extracted structured light features and intensity features are correlated to obtain the profile features of the device monitoring point. This processing is repeated for each set of light field distribution information and reflected light intensity information to obtain multiple profile features (e.g., five profile features). These profile features are superimposed to obtain the final profile features of the device monitoring point.

[0073] In this embodiment, by superimposing the contour features determined each time, a more accurate contour feature of the equipment monitoring point can be obtained, which can better reflect the contours of each object in the equipment monitoring point. If there is a leak at the equipment monitoring point, this contour feature can clearly reflect it.

[0074] S1024: Determine monitoring results based on the contour features.

[0075] Since the contour feature can accurately reflect the contour of each object in the equipment monitoring point, it is possible to detect whether there is a liquid contour feature in the contour feature, that is, to detect whether there is a liquid contour in the contour of each object reflected by the contour feature.

[0076] If the contour features are detected to contain liquid contours, that is, if the contour features reflect the contours of the objects, the monitoring result is determined to be abnormal. If the contour features are not detected to contain liquid contours, that is, if the contour features reflect the contours of the objects, the monitoring result is determined to be normal.

[0077] In this embodiment, the leakage monitoring model extracts structured light features from the light field distribution information and intensity features from the reflected light intensity information. Based on these features, the contour features of the device monitoring point are determined. This contour feature accurately reflects the outlines of each object at the device monitoring point. Therefore, based on this contour feature, it is possible to accurately determine whether the current device monitoring point is in a normal or abnormal state.

[0078] See Figure 4 , Figure 4 FIG. 1 is a schematic flow chart of a method for monitoring hydropower station equipment shown in another exemplary embodiment of the present application; FIG. Figure 4 The method for monitoring hydropower station equipment shown may include: S201 to S205. Figure 1 S101 to S102 are the same as those shown in FIG. For details, please refer to the description of S101 to S102 and will not be repeated here. S203 to S205 are as follows:

[0079] S203: When the monitoring result indicates that the device monitoring point is in an abnormal state, a ghost imaging image corresponding to the device monitoring point is generated according to the light field distribution information and the reflected light intensity information.

[0080] The monitoring result includes whether the equipment monitoring point is in a normal state or an abnormal state. When the monitoring result shows that the equipment monitoring point is in a normal state, the state of the equipment monitoring point continues to be monitored, that is, whether there is leakage at the equipment monitoring point.

[0081] When the monitoring result indicates that the device monitoring point is in an abnormal state, the collected reflected light intensity information and light field distribution information are correlated to obtain the ghost imaging image corresponding to the device monitoring point. The specific correlation calculation process can be referred to the existing technology and will not be repeated here.

[0082] S204: Compare the ghost imaging image with the preset liquid leakage image.

[0083] The preset leakage image is a pre-captured image of a leaking device at the monitoring point. For example, a leaking device at the monitoring point is simulated and captured to create a corresponding leakage image. The same method can be used for different monitoring points to generate corresponding leakage images for each device, and each corresponding leakage image is stored in the database.

[0084] A preset liquid leakage image corresponding to the ghost image is searched in a database, and the ghost image is compared with the preset liquid leakage image. Exemplarily, a similarity between the ghost image and the preset liquid leakage image is calculated. For example, the cosine similarity between the ghost image and the preset liquid leakage image is calculated using a cosine distance formula.

[0085] S205: Determine the final monitoring result of the part to be monitored based on the comparison result.

[0086] Exemplarily, the comparison result includes whether the ghost image is similar to a preset liquid leakage image, or whether the ghost image is dissimilar to the preset liquid leakage image. A similarity threshold is preset, and the calculated similarity between the ghost image and the preset liquid leakage image is compared with the preset similarity threshold. When the similarity is greater than or equal to the similarity threshold, the comparison result is determined to be similar to the preset liquid leakage image. When the similarity is less than the similarity threshold, the comparison result is determined to be dissimilar to the preset liquid leakage image.

[0087] If the comparison result shows that the ghost imaging image is not similar to the preset leakage image, the final monitoring result of the monitored portion is determined to be leakage at the device monitoring point. If the comparison result shows that the ghost imaging image is similar to the preset leakage image, the final monitoring result of the monitored portion is determined to be no leakage at the device monitoring point.

[0088] In this embodiment, when the leakage monitoring model outputs a monitoring result indicating an abnormal state at a device monitoring point, the ghost image is further compared with the preset leakage image. Based on the comparison result, it is re-determined whether the device monitoring point has a leak, thereby determining the final monitoring result for the device monitoring point. This provides dual monitoring protection, avoids false detections, and improves the accuracy of the monitoring results.

[0089] Optionally, in some possible implementations of the present application, the above S205 may further include S206, which is as follows:

[0090] S206: When the final monitoring result indicates that there is leakage at the equipment monitoring point, an early warning reminder is issued.

[0091] For example, when the final monitoring result shows that the equipment monitoring point has no leakage, the state of the equipment monitoring point is continuously monitored, that is, whether the equipment monitoring point has leakage. When the final monitoring result shows that the equipment monitoring point has leakage, an early warning reminder is issued.

[0092] For example, when leakage is detected at an equipment monitoring point, an early warning can be issued through any method or any combination of text, voice, video, picture, etc., so that the user can promptly understand the current leakage situation of the equipment monitoring point, facilitate timely resolution of the leakage problem at the equipment monitoring point, and thus ensure the safety of the hydropower station equipment.

[0093] In this embodiment, when it is determined that the equipment monitoring point is leaking, a timely warning is issued, which allows users to promptly understand the current leakage situation of the equipment monitoring point, and is conducive to users to promptly solve the problem of leakage at the equipment monitoring point, thereby ensuring the safety of hydropower station equipment.

[0094] Optionally, in some possible implementations of the present application, the above S206 may include S2061 to S2062, which are specifically as follows:

[0095] S2061: When the final monitoring result is that the equipment monitoring point is leaking, the leakage level of the equipment monitoring point is determined based on the ghost imaging image.

[0096] For example, when the final monitoring result is that there is leakage at the equipment monitoring point, the size of the leakage area is determined based on the ghost imaging image, and then the leakage level of the equipment monitoring point is determined based on the size of the leakage area.

[0097] For example, a neural network model is obtained from the network. The neural network model can segment the leakage region in the ghost imaging image and measure the size of the leakage region to obtain the size of the leakage area. For example, different leakage region sizes are pre-set to correspond to different leakage area sizes. The leakage area size is determined based on the measured size of the leakage region.

[0098] Different leakage area ranges are pre-set to correspond to different leakage levels. For example, the first leakage area range corresponds to leakage level 1; the second leakage area range corresponds to leakage level 2; and the third leakage area range corresponds to leakage level 3. The larger the leakage area, the higher the leakage level, i.e., the more serious the leakage. This is for illustrative purposes only and is not intended to be limiting.

[0099] For example, after the size of the leakage area is determined based on the ghost imaging image, it is determined to which leakage area range the leakage area belongs, and then the leakage level of the equipment monitoring point is determined based on the leakage level corresponding to the leakage area range.

[0100] S2062: Determine the warning method according to the leakage level, and issue a warning according to the warning method.

[0101] For example, different leakage levels are pre-set to correspond to different warning methods, and each leakage level and its corresponding warning method are stored in a database. After the leakage level of the current equipment monitoring point is determined based on the ghost imaging image in S2061, the warning method corresponding to the leakage level is searched in the database and a warning is issued using the warning method.

[0102] For example, the leakage level can be level one, level two, or level three, and the warning method can be text warning, voice warning, or voice and picture warning. Among them, when the leakage level is level one, the corresponding warning method is text warning (such as pre-edited warning information). For example, when the leakage level is determined to be level one, the warning information is sent to the user through email, text message, etc. The warning information contains the identification information of the equipment monitoring point and the specific location where the leakage occurs. After receiving the warning information, the user knows that a leakage has occurred at the current equipment monitoring point, but it is not very serious. The user can reasonably arrange a way to solve the leakage according to his or her own work, thereby ensuring the safety of the hydropower station equipment.

[0103] For example, when the leakage level is determined to be level 2, a voice warning is issued. For example, when the leakage level is determined to be level 2, an alarm sounds near the equipment monitoring point. When the user hears the alarm, he or she knows that a slightly serious leakage has occurred at the current equipment monitoring point, and can promptly resolve the leakage problem at the equipment monitoring point, thereby ensuring the safety of the hydropower station equipment.

[0104] For another example, when the leakage level is determined to be level three, a voice and image warning is issued. For example, when the leakage level is determined to be level three, an alarm sounds near the equipment monitoring point, and a ghost image is sent to the user via email, text message, etc. Upon hearing the alarm and receiving the ghost image, the user knows that a serious leakage has occurred at the current equipment monitoring point and can immediately resolve the leakage problem, thereby ensuring the safety of the hydropower station equipment.

[0105] In this embodiment, different early warning methods are adopted for different leakage levels. Users can instantly judge whether the leakage at the current equipment monitoring point is serious based on the early warning method received, and thus reasonably arrange ways to solve the leakage, thereby ensuring the safety of hydropower station equipment.

[0106] See Figure 5 , Figure 5 This is a schematic flow chart of a method for training a leakage monitoring model provided by the present application. The method for training a leakage monitoring model can be performed before step S101. Figure 5 The method for training a leakage monitoring model may include steps S301 to S305, specifically as follows:

[0107] S301: Obtain a sample training set.

[0108] The sample training set includes a plurality of sample light field distribution information, sample reflected light intensity information corresponding to each sample light field distribution information, and sample monitoring results corresponding to each sample light field distribution information.

[0109] For example, the sample light field distribution information and sample reflected light intensity information corresponding to different sample device monitoring points can be collected in advance, and the sample monitoring results corresponding to the sample light field distribution information can be marked according to the actual situation of the current sample device monitoring point, that is, the sample monitoring results corresponding to the sample device monitoring point.

[0110] For the same sample device monitoring point, one set of sample light field distribution information and sample reflected light intensity information may be collected, or multiple sets of sample light field distribution information and sample reflected light intensity information may be collected, without limitation.

[0111] Optionally, for the same sample device monitoring point, a scenario when the sample device monitoring point is leaking and a scenario when the sample device monitoring point is not leaking can be set, and one or more groups of sample light field distribution information and sample reflected light intensity information under different scenarios can be collected respectively, and the corresponding sample monitoring results can be marked.

[0112] The sample monitoring result is a monitoring result of a sample device monitoring point corresponding to the sample light field distribution information. The sample monitoring result may include that the sample device monitoring point is in a normal state or that the sample device monitoring point is in an abnormal state.

[0113] S302: Inputting the sample light field distribution information and the sample reflected light intensity information in the sample training set into the initial model for processing to obtain actual monitoring results of the sample light field distribution information.

[0114] For example, the network structure corresponding to the initial model during training is identical to the network results corresponding to the actual use of the leakage monitoring model. Therefore, the process of processing the sample light field distribution information and the sample reflected light intensity information using the initial model to obtain the actual monitoring results of the sample light field distribution information can be referred to the detailed description of S102 above and will not be repeated here.

[0115] It is worth noting that the sample light field distribution information and the sample reflected light intensity information are obtained by collecting the sample device monitoring point, and the actual monitoring result of the sample light field distribution information output by the initial model is the actual monitoring result of the sample device monitoring point.

[0116] S303: Calculate the loss value between the actual monitoring result and the sample monitoring result corresponding to the sample light field distribution information according to a preset loss function.

[0117] The loss value between the actual monitoring result and the sample monitoring result corresponding to the sample light field distribution information is used to measure the accuracy of the monitoring result. In this example, the preset loss function can be a cross-entropy loss function, a mean square error formula, etc. This loss function is used to calculate the loss value between the actual monitoring result and the sample monitoring result corresponding to the sample light field distribution information.

[0118] When a loss value is obtained between the actual monitoring result and the sample monitoring result corresponding to the sample light field distribution information, the difference between the loss value and the preset threshold is determined. If the loss value is greater than the preset threshold, S304 is executed; if the loss value is less than or equal to the preset threshold, S305 is executed.

[0119] S304: When it is detected that the loss value is greater than a preset threshold, the model parameters of the initial model are adjusted, and the initial model is continued to be trained using the sample training set.

[0120] Determine the difference between the loss value and a preset threshold. When the device executing the training process (e.g., this device or another device) determines that the current loss value is greater than the preset threshold, it determines that the initial model currently being trained does not meet the requirements. At this point, the model parameters (such as weight values) of the initial model being trained are adjusted, and the initial model is continued to be trained using the sample training set.

[0121] S305: When it is detected that the loss value is less than or equal to the preset threshold, the training of the initial model is stopped, and the trained initial model is determined as the leakage monitoring model.

[0122] For example, when the device performing the training process (e.g., this device or another device) determines that the current loss value is less than or equal to a preset threshold, it is determined that the initial model currently being trained meets the requirements. At this time, the model parameters in the initial model are fixed, and the initial model after the fixed model parameters is determined to be the trained leakage monitoring model.

[0123] In this embodiment, the leakage monitoring model is trained using multiple sample light field distribution information, the sample reflected light intensity information corresponding to each sample light field distribution information, and the sample monitoring results corresponding to each sample light field distribution information to obtain an initial model. During the training process, the model learns how to output accurate monitoring results based on the sample light field distribution information and sample reflected light intensity information. Therefore, in actual implementation, accurate monitoring results for each device monitoring point can be obtained by processing the light field distribution information and reflected light intensity information of the device monitoring point using the leakage monitoring model.

[0124] Optionally, in a possible implementation, the input and output data of the leakage monitoring model during actual use may be used to update the sample training set, and the updated sample training set may be used to continue training the leakage monitoring model.

[0125] For example, the input data of the leakage monitoring model during actual use (such as the light field distribution information and reflected light intensity information collected at the device monitoring point using ghost imaging technology) and the output data of the leakage monitoring model during actual use (such as the device monitoring point is in a normal state or the device monitoring point is in an abnormal state) are added to the original sample training set, and the leakage monitoring model is further trained using the sample training set with the added data. This continuous updating of the leakage monitoring model can ensure the accuracy of the monitoring results output by the leakage monitoring model.

[0126] See Figure 6 , Figure 6 This is a schematic diagram of a device for monitoring hydropower station equipment provided by an embodiment of the present application. The device for monitoring hydropower station equipment includes various units for performing Figure 1 、 Figures 3 to 5 Each step in the corresponding embodiment. Please refer to Figure 1 、 Figures 3 to 5 For the convenience of explanation, only the parts related to this embodiment are shown. Figure 6 ,include:

[0127] The acquisition unit 410 is used to acquire light field distribution information and reflected light intensity information of the device monitoring point using ghost imaging technology;

[0128] The processing unit 420 is used to input the light field distribution information and the reflected light intensity information into a trained leakage monitoring model for processing to obtain the monitoring results of the equipment monitoring point. The leakage monitoring model is obtained by training the initial model using a sample training set.

[0129] Optionally, the processing unit 420 is specifically configured to:

[0130] extracting structured light features of the light field distribution information through the liquid leakage monitoring model;

[0131] extracting the intensity characteristics of the reflected light intensity information through the leakage monitoring model;

[0132] determining a contour feature of the device monitoring point according to the structured light feature and the intensity feature;

[0133] The monitoring result is determined based on the contour feature.

[0134] Optionally, the monitoring result includes that the equipment monitoring point is in a normal state or that the equipment monitoring point is in an abnormal state, and the apparatus further includes:

[0135] a generating unit, configured to generate a ghost imaging image corresponding to the device monitoring point according to the light field distribution information and the reflected light intensity information when the monitoring result indicates that the device monitoring point is in an abnormal state;

[0136] a comparing unit, configured to compare the ghost imaging image with a preset liquid leakage image;

[0137] The determination unit is used to determine the final monitoring result of the part to be monitored according to the comparison result.

[0138] Optionally, the device further comprises a training unit, configured to:

[0139] Acquire the sample training set, where the sample training set includes a plurality of sample light field distribution information, sample reflected light intensity information corresponding to each of the sample light field distribution information, and a sample monitoring result corresponding to each of the sample light field distribution information;

[0140] Inputting the sample light field distribution information and the sample reflected light intensity information in the sample training set into the initial model for processing to obtain actual monitoring results of the sample light field distribution information;

[0141] Calculating a loss value between the actual monitoring result and the sample monitoring result corresponding to the sample light field distribution information according to a preset loss function;

[0142] When it is detected that the loss value is greater than a preset threshold, adjusting the model parameters of the initial model and continuing to train the initial model using the sample training set;

[0143] When it is detected that the loss value is less than or equal to the preset threshold, the training of the initial model is stopped, and the trained initial model is determined as the leakage monitoring model.

[0144] Optionally, the collection unit 410 is specifically configured to:

[0145] Obtain initial monitoring results of the leakage sensor;

[0146] When the initial monitoring result indicates that the equipment monitoring point is abnormal, the light field distribution information and the reflected light intensity information are collected using a ghost imaging method.

[0147] Optionally, the device further comprises:

[0148] The early warning unit is used to issue an early warning reminder when the final monitoring result is that the equipment monitoring point is leaking.

[0149] Optionally, the early warning unit is specifically configured to:

[0150] When the final monitoring result indicates that the equipment monitoring point is leaking, determining the leakage level of the equipment monitoring point according to the ghost imaging image;

[0151] An early warning method is determined according to the leakage level, and an early warning is performed according to the early warning method.

[0152] See Figure 7 , Figure 7 This is a schematic diagram of a terminal device provided by another embodiment of the present application. Figure 7 As shown, the terminal device 5 of this embodiment includes: a processor 50, a memory 51, and a computer program 52 stored in the memory 51 and executable on the processor 50. When the processor 50 executes the computer program 52, the steps in the above-mentioned various embodiments of the method for monitoring hydropower station equipment are implemented, such as Figure 1 Alternatively, when the processor 50 executes the computer program 52, the functions of the units in the above embodiments are realized, for example, Figure 6 The functions of units 410 to 420 are shown.

[0153] Exemplarily, the computer program 52 may be divided into one or more units, which are stored in the memory 51 and executed by the processor 50 to implement the present application. The one or more units may be a series of computer instruction segments capable of performing specific functions, which describe the execution process of the computer program 52 in the terminal device 5. For example, the computer program 52 may be divided into an acquisition unit and a processing unit, with the specific functions of each unit being as described above.

[0154] The terminal device may include, but is not limited to, a processor 50 and a memory 51. Those skilled in the art will understand that Figure 7 It is only an example of the terminal device 5 and does not constitute a limitation on the terminal device. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include input and output devices, network access devices, buses, etc.

[0155] The processor 50 may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0156] The memory 51 may be an internal storage unit of the terminal device, such as a hard disk or memory of the terminal device. The memory 51 may also be an external storage terminal of the terminal device, such as a plug-in hard disk equipped with the terminal device, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Furthermore, the memory 51 may include both an internal storage unit of the terminal device and an external storage terminal. The memory 51 is used to store the computer instructions and other programs and data required by the terminal. The memory 51 may also be used to temporarily store data that has been output or is about to be output.

[0157] An embodiment of the present application also provides a computer storage medium, which can be non-volatile or volatile. The computer storage medium stores a computer program, which, when executed by a processor, implements the steps in the above-mentioned method embodiments for monitoring hydropower station equipment.

[0158] The present application also provides a computer program product. When the computer program product is run on a terminal device, the terminal device executes the steps in the above-mentioned embodiments of the method for monitoring hydropower station equipment.

[0159] An embodiment of the present application also provides a chip or integrated circuit, which includes: a processor for calling and running a computer program from a memory, so that a device equipped with the chip or integrated circuit executes the steps in the above-mentioned method embodiments for monitoring hydropower station equipment.

[0160] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0161] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0162] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0163] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit of the technical solutions of the embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for monitoring hydropower station equipment, characterized in that: include: Using ghost imaging technology to collect light field distribution information and reflected light intensity information at the device monitoring point; the light field distribution information includes the position information and direction information of each object at the device monitoring point, and the reflected light intensity information includes the total light intensity reflected by each object at the device monitoring point; The structured light features of the light field distribution information are extracted using a trained leakage monitoring model, and the structured light features are used to represent the features of each object at the device monitoring point after being irradiated by light. The intensity features of the reflected light intensity information are extracted using the leakage monitoring model, and the intensity features are used to represent the intensity values ​​generated by each object at the device monitoring point after being irradiated by light. The structured light features and the intensity features are associated to obtain the contour features of the device monitoring point, and the contour features are used to map the contours of each object at the device monitoring point. The monitoring results are determined based on the contour features. The leakage monitoring model is obtained by training an initial model using a sample training set. When the monitoring result shows that the device monitoring point is in an abnormal state, the light field distribution information and the reflected light intensity information are correlated to generate a ghost imaging image corresponding to the device monitoring point; the ghost imaging image is compared with a preset leakage image; and the final monitoring result of the monitored part is determined based on the comparison result.

2. The method according to claim 1, wherein The method further comprises: Acquire the sample training set, where the sample training set includes a plurality of sample light field distribution information, sample reflected light intensity information corresponding to each of the sample light field distribution information, and a sample monitoring result corresponding to each of the sample light field distribution information; Inputting the sample light field distribution information and the sample reflected light intensity information in the sample training set into the initial model for processing to obtain actual monitoring results of the sample light field distribution information; Calculating a loss value between the actual monitoring result and the sample monitoring result corresponding to the sample light field distribution information according to a preset loss function; When it is detected that the loss value is greater than a preset threshold, adjusting the model parameters of the initial model and continuing to train the initial model using the sample training set; When it is detected that the loss value is less than or equal to the preset threshold, the training of the initial model is stopped, and the trained initial model is determined as the leakage monitoring model.

3. The method according to claim 1, wherein The method of collecting light field distribution information and reflected light intensity information of the monitoring point of the device using ghost imaging technology includes: Obtain initial monitoring results of the leakage sensor; When the initial monitoring result indicates that the equipment monitoring point is abnormal, the light field distribution information and the reflected light intensity information are collected using a ghost imaging method.

4. The method according to claim 1, wherein The final monitoring result includes whether the equipment monitoring point is leaking or not leaking. After determining the final monitoring result of the part to be monitored based on the comparison result, the method further includes: When the final monitoring result indicates that there is leakage at the monitoring point of the equipment, an early warning reminder is issued.

5. The method according to claim 4, wherein When the final monitoring result indicates that the equipment monitoring point is leaking, issuing an early warning reminder includes: When the final monitoring result indicates that the equipment monitoring point is leaking, determining the leakage level of the equipment monitoring point according to the ghost imaging image; An early warning method is determined according to the leakage level, and an early warning is performed according to the early warning method.

6. A device for monitoring hydropower station equipment, characterized in that: include: An acquisition unit, configured to acquire light field distribution information and reflected light intensity information of a monitoring point of the device using ghost imaging technology; The light field distribution information includes the position information and direction information of each object at the device monitoring point, and the reflected light intensity information includes the total light intensity reflected by each object at the device monitoring point; a processing unit configured to extract structured light features of the light field distribution information using a trained leakage monitoring model, the structured light features being used to represent features of each object at the device monitoring point after being irradiated by light; extract intensity features of the reflected light intensity information using the leakage monitoring model, the intensity features being used to represent intensity values ​​generated by each object at the device monitoring point after being irradiated by light; associate the structured light features with the intensity features to obtain contour features of the device monitoring point, the contour features being used to map the contours of each object at the device monitoring point; and determine the monitoring result based on the contour features; the leakage monitoring model being obtained by training an initial model using a sample training set; When the monitoring result shows that the device monitoring point is in an abnormal state, the light field distribution information and the reflected light intensity information are correlated to generate a ghost imaging image corresponding to the device monitoring point; the ghost imaging image is compared with a preset leakage image; and the final monitoring result of the monitored part is determined based on the comparison result.

7. A terminal device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images