Intelligent inspection method, system and equipment for intelligent power plant

Abstract

The invention provides an intelligent inspection method, system and device for a smart power plant, and relates to the technical field of smart energy, and the method comprises the steps: collecting the real-time equipment data of target equipment in a target smart power plant, constructing a three-dimensional virtual power station model, obtaining the historical and real-time operation data of the target equipment, constructing an equipment fault prediction model based on a machine learning algorithm to locate a fault position, collecting a photovoltaic inspection image of a photovoltaic module in the target intelligent power station through unmanned inspection equipment, carrying out hot spot recognition to generate potential photovoltaic hot spot information, and recognizing an equipment surface abnormal state based on a visual technology and a deep learning algorithm. A fireproof detection network is constructed to monitor the current area temperature of a target intelligent power station, the current area temperature is compared with a preset temperature threshold value to generate abnormal temperature information, and finally an intelligent inspection result is generated and sent to a management end, so that efficient, accurate and real-time intelligent inspection of electric power facilities can be realized, the operation and maintenance efficiency is improved, and the operation and maintenance cost is reduced. And safe and stable operation of power facilities is ensured.

Description

Technical Field

[0001] The present invention relates to the field of smart energy technology, and in particular to an intelligent inspection method, system and equipment for a smart power plant. Background Art

[0002] With the expansion of the scale of new energy power stations and the improvement of operation and maintenance requirements, traditional inspection methods can no longer meet the needs of safe and stable operation of power facilities.

[0003] At present, the operation and maintenance of power facilities mainly rely on manual inspections and simple video surveillance systems based on fixed cameras. These methods are often inefficient and difficult to achieve refined management and predictive maintenance, which directly affects the safe and stable operation of power facilities. In addition, traditional inspection methods have many defects. Specifically, manual inspection methods are often time-consuming and labor-intensive, easily affected by bad weather, and unable to achieve real-time monitoring; while simple video surveillance systems based on fixed cameras may have defects such as limited coverage, lack of autonomous analysis capabilities and active early warning mechanisms.

[0004] Therefore, it is necessary to provide an intelligent inspection method, system and equipment for a smart power plant to solve the above technical problems. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides an intelligent inspection method, system and equipment for a smart power plant to solve the problems of inefficiency, untimeliness, insufficient accurate monitoring and fault warning, and difficulty in early detection of fire hazards under traditional inspection methods.

[0006] The present invention provides an intelligent inspection method for a smart power plant, the inspection method comprising: Collecting real-time device data of target devices in a target smart power station, and constructing a three-dimensional virtual power station model corresponding to the target smart power station according to the real-time device data; Acquire historical operation data and real-time operation data of the target device, and build a device failure prediction model corresponding to the target device based on a machine learning algorithm to identify and locate the device failure location of the target device; Collecting photovoltaic inspection images of photovoltaic components in the target smart power station through unmanned inspection equipment, and performing hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spot information; Identify abnormal surface conditions of the target device based on visual technology and deep learning algorithms; Constructing a fire detection network to monitor the current regional temperature of the target smart power station and compare it with a preset temperature threshold to generate corresponding abnormal temperature information; Based on the equipment fault location, potential photovoltaic hot spot information, surface abnormal state and abnormal temperature information, combined with the three-dimensional virtual power station model, an intelligent inspection result of the target smart power station is generated and sent to the management end.

[0007] Preferably, the collecting of real-time device data of a target device in a target smart power station and constructing a three-dimensional virtual power station model corresponding to the target smart power station according to the real-time device data specifically includes: An interpolation surface at each data point in the real-time device data is determined based on a radial basis function. The expression of the interpolation surface at the current data point position in the real-time device data is as follows: In the formula, Represents the current data point position in real-time device data; Represents the interpolation surface at the current data point position x; Indicates the number of data points adjacent to the current data point; represents the three-dimensional interpolation coefficient; represents radial basis function; Indicates the position of the mth data point adjacent to the current data point; Indicates the current data point position x and the mth data point position The distance between represents the interpolation stability coefficient at the current data point position x; Σ represents the summation symbol; After determining the interpolation surfaces at the positions of all data points in the real-time device data, a three-dimensional virtual power station model corresponding to the target smart power station is constructed in combination with the real-time device data.

[0008] Preferably, after obtaining the three-dimensional virtual power station model corresponding to the target smart power station, the three-dimensional virtual power station model is optimized accordingly by adjusting the position of each data point in the real-time device data based on a network optimization algorithm, and the expression of the adjusted current data point position in the real-time device data is as follows: In the formula, Represents the current data point position in the adjusted real-time device data; Represents the current data point position in real-time device data; represents the smoothing strength coefficient; Indicates the number of data points adjacent to the current data point; represents the mth data point adjacent to the current data point; Σ represents the summation symbol; According to the adjusted position of each data point in the real-time equipment data, the optimized three-dimensional virtual power station model is obtained.

[0009] Preferably, the acquiring of historical operation data and real-time operation data of the target device, and building a device failure prediction model corresponding to the target device based on a machine learning algorithm, and identifying and locating the device failure location of the target device, specifically includes: Acquire historical operation data and real-time operation data of the target device, extract key features from the historical operation data to generate corresponding historical operation features, and extract key features from the real-time operation data to generate corresponding real-time operation features; Building a device failure prediction model corresponding to the target device based on the historical operation characteristics and the real-time operation characteristics based on a machine learning algorithm; Fault detection is performed on the target device based on the device fault prediction model to identify and locate the device fault location of the target device.

[0010] Preferably, the collecting of photovoltaic inspection images of photovoltaic modules in the target smart power station by unmanned inspection equipment, and performing hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spot information specifically includes: Planning a preset inspection route for the unmanned inspection equipment according to the distribution of photovoltaic components in the target smart power station; Controlling the unmanned inspection device to perform inspection according to the preset inspection route, and photographing the photovoltaic components in the target smart power station through the inspection sensor carried by the unmanned inspection device to generate the photovoltaic inspection image; Feature extraction is performed on the photovoltaic inspection image to generate photovoltaic inspection features, and hot spot identification is performed on the photovoltaic inspection features to generate the potential photovoltaic hot spot information.

[0011] Preferably, the extracting features of the photovoltaic inspection image to generate photovoltaic inspection features, and identifying hot spots on the photovoltaic inspection features to generate the potential photovoltaic hot spot information specifically include: The photovoltaic inspection image is grayed, and feature extraction is performed on the photovoltaic inspection image after graying to generate the photovoltaic inspection feature. The expression of the photovoltaic inspection feature is: In the formula, represents the photovoltaic inspection feature; i and j represent the position of the characteristic pixel points in the photovoltaic inspection feature; Neighborhood standard deviation of characteristic pixels representing photovoltaic inspection characteristics; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i, j; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position j.

[0012] Preferably, the expression for calculating the extreme value of the photovoltaic inspection characteristic is as follows: In the formula, Indicates photovoltaic inspection characteristics; represents the extreme value of the photovoltaic inspection feature; i and j represent the position of the characteristic pixel points in the photovoltaic inspection feature; Neighborhood standard deviation of characteristic pixels representing photovoltaic inspection characteristics; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i, j; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position j; The characteristic pixel point corresponding to the extreme value of the photovoltaic inspection feature is selected as the stable photovoltaic inspection feature point, the photovoltaic inspection feature direction is determined according to the wavelet feature of the circular neighborhood of the stable photovoltaic inspection feature point, and the 4*4 rectangular area corresponding to the stable photovoltaic inspection feature point is obtained. The sum of the horizontal direction value, the sum of the vertical direction value, the sum of the horizontal direction absolute value and the sum of the vertical direction absolute value of the stable photovoltaic inspection feature point are counted to obtain the potential photovoltaic hot spot feature, and the potential photovoltaic hot spot information corresponding to the photovoltaic inspection image is determined according to the potential photovoltaic hot spot feature.

[0013] An intelligent inspection system for a smart power plant, applied to an intelligent inspection method for a smart power plant according to any one of claims 1 to 7, the inspection system comprising: A model building module, used to collect real-time device data of target devices in a target smart power station, and build a three-dimensional virtual power station model corresponding to the target smart power station according to the real-time device data; A fault location module is used to obtain historical operation data and real-time operation data of the target device, and to construct a device fault prediction model corresponding to the target device based on a machine learning algorithm to identify and locate the device fault location of the target device; A hot spot identification module is used to collect photovoltaic inspection images of photovoltaic modules in the target smart power station through unmanned inspection equipment, and perform hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spot information; A surface anomaly module, used to identify the surface abnormality state of the target device based on visual technology and deep learning algorithm; A temperature comparison module is used to build a fire detection network to monitor the current regional temperature of the target smart power station and compare it with a preset temperature threshold to generate corresponding abnormal temperature information; The result sending module is used to generate the intelligent inspection result of the target smart power station and send it to the management end based on the equipment fault location, potential photovoltaic hot spot information, surface abnormal state and abnormal temperature information, combined with the three-dimensional virtual power station model.

[0014] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the intelligent inspection method of a smart power plant as described in any one of the above items.

[0015] Compared with the related art, the intelligent inspection method, system and equipment of a smart power plant provided by the present invention have the following beneficial effects: The present invention can collect real-time equipment data of target equipment in the target smart power station through the model building module, fault identification module, hot spot identification module, surface anomaly module, temperature comparison module and result sending module, and build a three-dimensional virtual power station model corresponding to the target smart power station according to the real-time equipment data; obtain the historical operation data and real-time operation data of the target equipment, and build an equipment fault prediction model corresponding to the target equipment based on the machine learning algorithm to identify and locate the equipment fault location of the target equipment; collect photovoltaic inspection images of photovoltaic components in the target smart power station through unmanned inspection equipment, and perform hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spot information; based on visual technology and deep learning algorithms Identify the abnormal surface state of the target equipment; build a fire detection network to monitor the current area temperature of the target smart power station and compare it with the preset temperature threshold to generate the corresponding abnormal temperature information; based on the equipment fault location, potential photovoltaic hot spot information, surface abnormal state and abnormal temperature information, combined with the three-dimensional virtual power station model, generate the intelligent inspection results of the target smart power station and send them to the management end, so as to realize efficient, accurate and real-time intelligent inspection of power facilities, improve the operation and maintenance efficiency of power facilities, reduce the operation and maintenance costs of power facilities, ensure their safe and stable operation, and solve the problems of inefficient, untimely and inaccurate monitoring and insufficient fault warning in the traditional inspection mode and difficulty in early detection of fire hazards.

[0016] The present invention can use machine learning algorithms to build equipment fault prediction models, accurately identify and locate equipment fault locations, thereby achieving efficient, accurate, and real-time intelligent inspections of power facilities, improving the accuracy of fault predictions, significantly reducing unexpected downtime events, and improving equipment availability and operation and maintenance quality; at the same time, the present invention can greatly reduce the cost and time consumption of manual inspections by applying unmanned intelligent inspection technology, and improve inspection efficiency and accuracy; the present invention can monitor the regional temperature of the power station in real time and identify abnormal temperature conditions of the power station by building a fire detection network, and then can timely detect and warn of potential fire risks in the power station, greatly reducing property losses caused by fire, ensuring the continuity and safety of power station operations, and ensuring the safe and stable operation of power facilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flow chart of an intelligent inspection method for a smart power plant provided by an embodiment of the present invention; Figure 2 A system block diagram of an intelligent inspection system for a smart power plant provided by an embodiment of the present invention; Figure 3 A schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] like Figure 1 FIG. 1 is a flow chart of an intelligent inspection method for a smart power plant provided by an embodiment of the present invention. Figure 1 The execution subject of the method shown may be a software and / or hardware device. The execution subject of the present application may include but is not limited to at least one of the following: user equipment, network equipment, etc. Among them, user equipment may include but is not limited to computers, smart phones, personal digital assistants (PDA) and the electronic devices mentioned above. Network equipment may include but is not limited to a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers, wherein cloud computing is a type of distributed computing, a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. It includes steps S1 to S6, as follows: S1, collecting real-time device data of target devices in a target smart power station, and constructing a three-dimensional virtual power station model corresponding to the target smart power station according to the real-time device data; Among them, the target smart power station refers to the power station that needs to be intelligently inspected; the target equipment refers to the equipment in the power station that needs to be intelligently inspected; the real-time equipment data includes the layout data and working status data of the power station equipment collected in real time; the three-dimensional virtual power station model refers to the three-dimensional virtual model corresponding to the actual power station.

[0020] It is understandable that the sensor network deployed in the power station can collect the layout data and working status data of the power station equipment in real time, such as equipment location, equipment working voltage, equipment working current, equipment working temperature, etc. Using these data, combined with 3D modeling technology, a 3D virtual power station model that is highly consistent with the physical power station can be constructed.

[0021] S2, obtaining historical operation data and real-time operation data of the target device, and building a device failure prediction model corresponding to the target device based on a machine learning algorithm to identify and locate the device failure location of the target device; Among them, historical operation data refers to the operating status data of power station equipment in the past stage; real-time operation data refers to the real-time operating status data of power station equipment in the current stage; equipment failure prediction model refers to the model for predicting and judging the failure conditions of power station equipment; equipment failure location refers to the location of the fault point on the power station equipment.

[0022] In practical applications, by integrating and analyzing the historical and real-time operating data of power plant equipment and using machine learning algorithms, such as time series analysis algorithms, a fault prediction model corresponding to the power plant equipment can be trained.

[0023] Through the above methods, potential fault points on power station equipment can be accurately identified and located, effectively preventing the occurrence of sudden equipment failures and improving the stability and safety of power station operation.

[0024] S3, collecting photovoltaic inspection images of photovoltaic modules in the target smart power station through unmanned inspection equipment, and performing hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spot information; Among them, photovoltaic inspection images refer to images of photovoltaic modules of power stations taken by unmanned inspection equipment during inspection; potential photovoltaic hot spot information refers to potential hot spot defect information on photovoltaic modules of power stations, specifically including hot spot location and hot spot type.

[0025] In actual applications, unmanned inspection equipment can be drones or unmanned inspection robots. Drones can fly autonomously or move to designated locations and take all-round photos of the photovoltaic components of the power station, while unmanned inspection robots can take close-up photos of the details on the sides of the photovoltaic components of the power station.

[0026] At certain specific locations, the images taken by drones and unmanned inspection robots can be stitched together to generate new images, namely the overall inspection images of the photovoltaic components of the power station. Then, these inspection images can be automatically analyzed using image processing technology and hot spot recognition algorithms to quickly identify potential photovoltaic hot spot information on the photovoltaic components of the power station to ensure efficient and stable operation of the power station.

[0027] S4, identifying the abnormal state of the surface of the target device based on visual technology and deep learning algorithm; It can be understood that surface abnormality refers to the abnormal state of the surface of power station equipment, including cracks, corrosion, dirt, etc. By integrating visual technology and deep learning algorithms, it is possible to accurately identify abnormal conditions such as cracks, corrosion, dirt, etc. on the surface of power station equipment, issue abnormal warnings in time, and notify staff to take corresponding maintenance measures.

[0028] S5, building a fire detection network to monitor the current regional temperature of the target smart power station and compare it with a preset temperature threshold to generate corresponding abnormal temperature information; Among them, the fire detection network refers to the sensor network for fire monitoring in power plants; the preset temperature threshold refers to the pre-set safe temperature of the power plant, that is, the maximum temperature that the power plant can withstand; the abnormal temperature information refers to the current temperature information of the power plant that exceeds the maximum temperature.

[0029] In practical applications, a fire detection network covering the entire station can be built, using infrared thermal imaging technology and intelligent sensors to monitor the temperature of each area of ​​the power station in real time and compare it with the preset power station safety temperature. Once abnormal temperature information is detected, the alarm mechanism will be immediately activated to ensure that the fire risk is controlled in a timely manner.

[0030] S6, based on the equipment fault location, potential photovoltaic hot spot information, surface abnormal state and abnormal temperature information, combined with the three-dimensional virtual power station model, generate the intelligent inspection result of the target smart power station and send it to the management end.

[0031] It can be understood that the management end refers to the terminal corresponding to the staff who manage the power station. Based on the collected equipment fault location, potential photovoltaic hot spot information, surface abnormal state and abnormal temperature information, combined with the three-dimensional virtual power station model, a detailed intelligent inspection report can be automatically generated and sent to the power station staff in real time through the network, so that the staff can respond quickly and take effective measures, thereby ensuring the continuous, stable and efficient operation of the power station.

[0032] Through the above methods, the inspection efficiency and accuracy of power stations can be greatly improved, and the cost and risk of manual inspection can be significantly reduced.

[0033] In a specific implementation process, the real-time device data of the target device in the target smart power station is collected, and a three-dimensional virtual power station model corresponding to the target smart power station is constructed according to the real-time device data, specifically including: An interpolation surface at each data point in the real-time device data is determined based on a radial basis function. The expression of the interpolation surface at the current data point position in the real-time device data is as follows: In the formula, Represents the current data point position in real-time device data; Represents the interpolation surface at the current data point position x; Indicates the number of data points adjacent to the current data point; represents the three-dimensional interpolation coefficient; represents radial basis function; Indicates the position of the mth data point adjacent to the current data point; Indicates the current data point position x and the mth data point position The distance between represents the interpolation stability coefficient at the current data point position x; Σ represents the summation symbol; After determining the interpolation surfaces at the positions of all data points in the real-time device data, a three-dimensional virtual power station model corresponding to the target smart power station is constructed in combination with the real-time device data.

[0034] It can be understood that by collecting the working status data of the power station equipment in real time and based on data analysis technology, a three-dimensional virtual model of the power station can be accurately constructed.

[0035] Specifically, the radial basis function interpolation method can be used to determine the corresponding interpolation surface for each data point in the collected real-time equipment data. After determining the interpolation surface at the position of all data points, the three-dimensional virtual model of the power station can be constructed using modeling technology and algorithms.

[0036] In the above way, by building a three-dimensional virtual power plant model, the actual power plant can be simulated and predicted in a virtual environment, thereby optimizing the operation and maintenance strategy and reducing the operation and maintenance cost. In addition, the three-dimensional virtual power plant model can provide data visualization function, intuitively display the structure and layout of the power plant, and facilitate the staff to deeply understand and analyze the power plant data and timely discover potential problems and risks.

[0037] After obtaining the three-dimensional virtual power station model corresponding to the target smart power station, based on the network optimization algorithm, the three-dimensional virtual power station model is optimized accordingly by adjusting the position of each data point in the real-time device data. The expression of the adjusted current data point position in the real-time device data is as follows: In the formula, Represents the current data point position in the adjusted real-time device data; Represents the current data point position in real-time device data; represents the smoothing strength coefficient; Indicates the number of data points adjacent to the current data point; represents the mth data point adjacent to the current data point; Σ represents the summation symbol; According to the adjusted position of each data point in the real-time equipment data, the optimized three-dimensional virtual power station model is obtained.

[0038] It is understandable that after obtaining the three-dimensional virtual power plant model corresponding to the power plant, the network optimization algorithm can be used to deeply optimize the model. Specifically, by adjusting the spatial position of each data point in the real-time device data set, it can be ensured that the three-dimensional virtual power plant model can more accurately reflect the operating status of the actual power plant.

[0039] Through the above method, the data visualization effect of the three-dimensional virtual power station model can be improved, which is convenient for subsequent power station operation analysis, fault diagnosis and intelligent scheduling.

[0040] The acquiring of historical operation data and real-time operation data of the target device, and building a device failure prediction model corresponding to the target device based on a machine learning algorithm, and identifying and locating the device failure position of the target device, specifically includes: Acquire historical operation data and real-time operation data of the target device, extract key features from the historical operation data to generate corresponding historical operation features, and extract key features from the real-time operation data to generate corresponding real-time operation features; Building a device failure prediction model corresponding to the target device based on the historical operation characteristics and the real-time operation characteristics based on a machine learning algorithm; Fault detection is performed on the target device based on the device fault prediction model to identify and locate the device fault location of the target device.

[0041] It should be noted that historical operation data and real-time operation data can be systematically obtained from power station equipment, thereby comprehensively analyzing the operation status of the power station equipment.

[0042] Specifically, key features can be accurately extracted from historical operation data. These features can reflect the performance and change trends of power station equipment in the past period of time, and then generate corresponding historical operation features. At the same time, the current real-time operation data can be synchronously processed to extract real-time operation features that can instantly reflect the operation status of the equipment.

[0043] Furthermore, machine learning algorithms can be used to combine historical and real-time operating characteristics to build a corresponding equipment failure prediction model for power station equipment. Through training and optimization, this model can learn the complex relationship between normal and abnormal equipment states, thereby achieving effective prediction of equipment failures.

[0044] Ultimately, based on the equipment fault prediction model, real-time fault detection and analysis can be performed on power station equipment. By comparing the model prediction results with the actual operating status of the equipment, the specific location of the equipment fault can be accurately identified and located, thereby effectively improving the efficiency and accuracy of power station equipment maintenance and facilitating subsequent fault detection and maintenance work.

[0045] The collecting of photovoltaic inspection images of photovoltaic modules in the target smart power station by unmanned inspection equipment and performing hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spot information specifically includes: Planning a preset inspection route for the unmanned inspection equipment according to the distribution of photovoltaic components in the target smart power station; Controlling the unmanned inspection device to perform inspection according to the preset inspection route, and photographing the photovoltaic components in the target smart power station through the inspection sensor carried by the unmanned inspection device to generate the photovoltaic inspection image; Feature extraction is performed on the photovoltaic inspection image to generate photovoltaic inspection features, and hot spot identification is performed on the photovoltaic inspection features to generate the potential photovoltaic hot spot information.

[0046] It can be understood that the photovoltaic inspection characteristics refer to the characteristic information of the photovoltaic component images captured by the unmanned inspection equipment during the inspection process.

[0047] Specifically, the inspection route of unmanned inspection equipment can be planned based on the specific distribution of photovoltaic modules in the power station, including the arrangement and spacing of modules and the overall layout of the power station, so as to ensure that the inspection equipment can fully cover all photovoltaic modules. At the same time, the inspection route can be optimized in real time, unnecessary inspection time can be reduced, and inspection efficiency can be improved.

[0048] Subsequently, intelligent control technology can be used to accurately direct the unmanned inspection equipment to conduct inspections along the preset inspection routes. During the inspection process, the high-precision inspection sensors on the unmanned inspection equipment have the characteristics of high sensitivity and high resolution, which can capture and accurately capture the surface information of photovoltaic modules in real time, generating clear and comprehensive inspection images.

[0049] Next, the collected inspection images can be processed for feature extraction, and image processing algorithms can be used to extract key features closely related to the status of photovoltaic modules, such as color, texture, brightness, etc., from the inspection images to form photovoltaic inspection features. These features can objectively reflect the current status of photovoltaic modules and facilitate subsequent hot spot identification.

[0050] Finally, the hot spot recognition algorithm can be used to conduct in-depth analysis of the PV inspection characteristics and identify potential PV hot spot information. Specifically, by comparing the characteristic differences between normal PV modules and abnormal modules, the location and severity of the modules with hot spot problems can be accurately identified, which is convenient for subsequent troubleshooting and repair work.

[0051] Through the above method, unmanned inspection equipment can be used to perform efficient and accurate inspections on photovoltaic components in power stations to identify and monitor potential photovoltaic hot spot problems, thereby improving the operation and maintenance efficiency of photovoltaic power stations and significantly reducing safety hazards and economic losses caused by hot spot problems.

[0052] The step of extracting features from the photovoltaic inspection image to generate photovoltaic inspection features, and performing hot spot identification on the photovoltaic inspection features to generate the potential photovoltaic hot spot information specifically includes: The photovoltaic inspection image is grayed, and feature extraction is performed on the photovoltaic inspection image after graying to generate the photovoltaic inspection feature. The expression of the photovoltaic inspection feature is: In the formula, represents the photovoltaic inspection feature; i and j represent the position of the characteristic pixel points in the photovoltaic inspection feature; Neighborhood standard deviation of characteristic pixels representing photovoltaic inspection characteristics; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i, j; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position j.

[0053] Specifically, the photovoltaic inspection image can first be grayed to simplify the image information and reduce the complexity of subsequent processing. Then, the key photovoltaic inspection features can be extracted using a feature extraction algorithm for the grayed inspection image. These features include the specific location of the feature pixel, the neighborhood standard deviation of the feature pixel, and the second-order derivative at the feature pixel location.

[0054] Through the above method, the photovoltaic inspection images can be gray-scaled and feature extracted, key information can be efficiently extracted from complex images, the real-time status of photovoltaic modules can be determined, and potential hot spot information can be effectively identified, which helps to accurately locate the hot spot position and assess its severity, significantly improving the operation and maintenance efficiency and safety of photovoltaic power stations.

[0055] The expression for calculating the extreme value of the photovoltaic inspection characteristic is as follows: In the formula, Indicates photovoltaic inspection characteristics; represents the extreme value of the photovoltaic inspection feature; i and j represent the position of the characteristic pixel points in the photovoltaic inspection feature; Neighborhood standard deviation of characteristic pixels representing photovoltaic inspection characteristics; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i, j; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position j; The characteristic pixel point corresponding to the extreme value of the photovoltaic inspection feature is selected as the stable photovoltaic inspection feature point, the photovoltaic inspection feature direction is determined according to the wavelet feature of the circular neighborhood of the stable photovoltaic inspection feature point, and the 4*4 rectangular area corresponding to the stable photovoltaic inspection feature point is obtained. The sum of the horizontal direction value, the sum of the vertical direction value, the sum of the horizontal direction absolute value and the sum of the vertical direction absolute value of the stable photovoltaic inspection feature point are counted to obtain the potential photovoltaic hot spot feature, and the potential photovoltaic hot spot information corresponding to the photovoltaic inspection image is determined according to the potential photovoltaic hot spot feature.

[0056] It can be understood that the extreme value of the photovoltaic inspection feature refers to the extreme value in these feature information, that is, the maximum or minimum value. The position of the feature pixel point can accurately locate the specific position of these extreme value features in the image. The neighborhood standard deviation of the feature pixel of the photovoltaic inspection feature reflects the degree of discreteness of the surrounding pixel values ​​of the feature pixel, which is an important indicator for measuring the stability of the feature. The second-order derivative of the photovoltaic inspection feature at the feature pixel point position is used to describe the rate of change of the image grayscale value at the feature pixel point, which helps to identify the edges and mutation points in the image.

[0057] Furthermore, the characteristic pixel points corresponding to the extreme values ​​can be selected as stable photovoltaic inspection feature points. Based on the wavelet features in the circular neighborhood of these stable feature points, the direction of the photovoltaic inspection feature can be determined. Subsequently, the 4*4 rectangular area corresponding to the stable feature point can be obtained, and the potential photovoltaic hot spot features can be extracted by counting the sum of the horizontal direction values, the sum of the vertical direction values, the sum of the horizontal absolute values, and the sum of the vertical absolute values ​​of the pixel points in the area. Finally, based on these potential photovoltaic hot spot features, the potential photovoltaic hot spot information corresponding to the photovoltaic inspection image can be further analyzed and determined.

[0058] like Figure 2 FIG. 1 is a system block diagram of an intelligent inspection system for a smart power plant provided by an embodiment of the present invention, wherein the inspection system includes: A model building module, used to collect real-time device data of target devices in a target smart power station, and build a three-dimensional virtual power station model corresponding to the target smart power station according to the real-time device data; A fault location module is used to obtain historical operation data and real-time operation data of the target device, and to construct a device fault prediction model corresponding to the target device based on a machine learning algorithm to identify and locate the device fault location of the target device; A hot spot identification module is used to collect photovoltaic inspection images of photovoltaic modules in the target smart power station through unmanned inspection equipment, and perform hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spot information; A surface anomaly module, used to identify the surface abnormality state of the target device based on visual technology and deep learning algorithm; A temperature comparison module is used to build a fire detection network to monitor the current regional temperature of the target smart power station and compare it with a preset temperature threshold to generate corresponding abnormal temperature information; The result sending module is used to generate the intelligent inspection result of the target smart power station and send it to the management end based on the equipment fault location, potential photovoltaic hot spot information, surface abnormal state and abnormal temperature information, combined with the three-dimensional virtual power station model.

[0059] Figure 2 The apparatus of the embodiment shown can be used to perform Figure 1 The implementation principles and technical effects of the steps in the method embodiment shown are similar and will not be repeated here.

[0060] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the intelligent inspection method of a smart power plant as described in any one of the above items.

[0061] like Figure 3FIG. 1 is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention. The electronic device 30 includes: a processor 31, a memory 32 and a computer program; wherein The memory 32 is used to store the computer program, which may also be a flash memory. The computer program is, for example, an application program, a functional module, etc. for implementing the above method.

[0062] The processor 31 is used to execute the computer program stored in the memory to implement each step performed by the device in the above method. For details, please refer to the relevant description in the above method embodiment.

[0063] Optionally, the memory 32 may be independent or integrated with the processor 31 .

[0064] When the memory 32 is a device independent of the processor 31, the device may further include: The bus 33 is used to connect the memory 32 and the processor 31 .

[0065] The present invention also provides a readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the methods provided by the various embodiments described above.

[0066] Among them, the readable storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application-specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). In addition, the ASIC can be located in a user device. Of course, the processor and the readable storage medium can also exist in a communication device as discrete components. The readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0067] The present invention also provides a program product, which includes an execution instruction, which is stored in a readable storage medium. At least one processor of a device can read the execution instruction from the readable storage medium, and at least one processor executes the execution instruction so that the device implements the methods provided in the above various embodiments.

[0068] In the embodiments of the above-mentioned devices, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0069] Through the introduction of the above embodiments, the present invention can collect real-time equipment data of target equipment in the target smart power station through the intelligent inspection method, system and equipment of the smart power plant, through the model building module, fault identification module, hot spot identification module, surface abnormality module, temperature comparison module and result sending module, and build a three-dimensional virtual power station model corresponding to the target smart power station according to the real-time equipment data; obtain the historical operation data and real-time operation data of the target equipment, and build an equipment fault prediction model corresponding to the target equipment based on the machine learning algorithm to identify and locate the equipment fault position of the target equipment; collect photovoltaic inspection images of photovoltaic components in the target smart power station through unmanned inspection equipment, and perform hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spots Information; Based on visual technology and deep learning algorithms, the abnormal surface state of the target equipment is identified; a fire detection network is built to monitor the current regional temperature of the target smart power station and compare it with the preset temperature threshold to generate the corresponding abnormal temperature information; based on the equipment fault location, potential photovoltaic hot spot information, surface abnormal state and abnormal temperature information, combined with a three-dimensional virtual power station model, the intelligent inspection results of the target smart power station are generated and sent to the management end, thereby realizing efficient, accurate and real-time intelligent inspection of power facilities, improving the operation and maintenance efficiency of power facilities, reducing the operation and maintenance costs of power facilities, and ensuring their safe and stable operation, solving the problems of inefficient, untimely and inaccurate monitoring and insufficient fault warning in the traditional inspection mode and difficulty in early detection of fire hazards.

[0070] The present invention can use machine learning algorithms to build equipment fault prediction models, accurately identify and locate equipment fault locations, thereby achieving efficient, accurate, and real-time intelligent inspections of power facilities, improving the accuracy of fault predictions, significantly reducing unexpected downtime events, and improving equipment availability and operation and maintenance quality; at the same time, the present invention can greatly reduce the cost and time consumption of manual inspections by applying unmanned intelligent inspection technology, and improve inspection efficiency and accuracy; the present invention can monitor the regional temperature of the power station in real time and identify abnormal temperature conditions of the power station by building a fire detection network, and then can timely detect and warn of potential fire risks in the power station, greatly reducing property losses caused by fire, ensuring the continuity and safety of power station operations, and ensuring the safe and stable operation of power facilities.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention 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 replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent inspection method for a smart power plant, characterized in that: The inspection method comprises: Collecting real-time device data of target devices in a target smart power station, and constructing a three-dimensional virtual power station model corresponding to the target smart power station according to the real-time device data; Acquire historical operation data and real-time operation data of the target device, and build a device failure prediction model corresponding to the target device based on a machine learning algorithm to identify and locate the device failure location of the target device; Collecting photovoltaic inspection images of photovoltaic components in the target smart power station through unmanned inspection equipment, and performing hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spot information; Identify abnormal surface conditions of the target device based on visual technology and deep learning algorithms; Constructing a fire detection network to monitor the current regional temperature of the target smart power station and compare it with a preset temperature threshold to generate corresponding abnormal temperature information; Based on the equipment fault location, potential photovoltaic hot spot information, surface abnormal state and abnormal temperature information, combined with the three-dimensional virtual power station model, an intelligent inspection result of the target smart power station is generated and sent to the management end.

2. The intelligent inspection method of a smart power plant according to claim 1, characterized in that: The collecting of real-time device data of a target device in a target smart power station and constructing a three-dimensional virtual power station model corresponding to the target smart power station according to the real-time device data specifically includes: An interpolation surface at each data point in the real-time device data is determined based on a radial basis function. The expression of the interpolation surface at the current data point position in the real-time device data is as follows: In the formula, Represents the current data point position in real-time device data; Represents the interpolation surface at the current data point position x; Indicates the number of data points adjacent to the current data point; represents the three-dimensional interpolation coefficient; represents radial basis function; Indicates the position of the mth data point adjacent to the current data point; Indicates the current data point position x and the mth data point position The distance between represents the interpolation stability coefficient at the current data point position x; Σ represents the summation symbol; After determining the interpolation surfaces at the positions of all data points in the real-time device data, a three-dimensional virtual power station model corresponding to the target smart power station is constructed in combination with the real-time device data.

3. The intelligent inspection method of a smart power plant according to claim 2, characterized in that: After obtaining the three-dimensional virtual power station model corresponding to the target smart power station, based on the network optimization algorithm, the three-dimensional virtual power station model is optimized accordingly by adjusting the position of each data point in the real-time device data. The expression of the adjusted current data point position in the real-time device data is as follows: In the formula, Represents the current data point position in the adjusted real-time device data; Represents the current data point position in real-time device data; represents the smoothing strength coefficient; Indicates the number of data points adjacent to the current data point; represents the mth data point adjacent to the current data point; Σ represents the summation symbol; According to the adjusted position of each data point in the real-time equipment data, the optimized three-dimensional virtual power station model is obtained.

4. The intelligent inspection method of a smart power plant according to claim 1, characterized in that: The acquiring of historical operation data and real-time operation data of the target device, and building a device failure prediction model corresponding to the target device based on a machine learning algorithm, and identifying and locating the device failure position of the target device, specifically includes: Acquire historical operation data and real-time operation data of the target device, extract key features from the historical operation data to generate corresponding historical operation features, and extract key features from the real-time operation data to generate corresponding real-time operation features; Building a device failure prediction model corresponding to the target device based on the historical operation characteristics and the real-time operation characteristics based on a machine learning algorithm; Fault detection is performed on the target device based on the device fault prediction model to identify and locate the device fault location of the target device.

5. The intelligent inspection method of a smart power plant according to claim 1, characterized in that: The collecting of photovoltaic inspection images of photovoltaic modules in the target smart power station by unmanned inspection equipment and performing hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spot information specifically includes: Planning a preset inspection route for the unmanned inspection equipment according to the distribution of photovoltaic modules in the target smart power station; Controlling the unmanned inspection device to perform inspection according to the preset inspection route, and photographing the photovoltaic components in the target smart power station through the inspection sensor carried by the unmanned inspection device to generate the photovoltaic inspection image; Feature extraction is performed on the photovoltaic inspection image to generate photovoltaic inspection features, and hot spot identification is performed on the photovoltaic inspection features to generate the potential photovoltaic hot spot information.

6. The intelligent inspection method of a smart power plant according to claim 5, characterized in that: The step of extracting features from the photovoltaic inspection image to generate photovoltaic inspection features, and performing hot spot identification on the photovoltaic inspection features to generate the potential photovoltaic hot spot information specifically includes: The photovoltaic inspection image is grayed, and feature extraction is performed on the photovoltaic inspection image after graying to generate the photovoltaic inspection feature. The expression of the photovoltaic inspection feature is: In the formula, represents the photovoltaic inspection feature; i and j represent the position of the characteristic pixel points in the photovoltaic inspection feature; Neighborhood standard deviation of characteristic pixels representing PV inspection characteristics; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i, j; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position j.

7. The intelligent inspection method of a smart power plant according to claim 6, characterized in that: The expression for calculating the extreme value of the photovoltaic inspection characteristic is as follows: In the formula, Indicates photovoltaic inspection characteristics; represents the extreme value of the photovoltaic inspection feature; i and j represent the position of the characteristic pixel points in the photovoltaic inspection feature; Neighborhood standard deviation of characteristic pixels representing PV inspection characteristics; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position i, j; Represents the second-order derivative of the photovoltaic inspection feature at the feature pixel position j; The characteristic pixel point corresponding to the extreme value of the photovoltaic inspection feature is selected as the stable photovoltaic inspection feature point, the photovoltaic inspection feature direction is determined according to the wavelet feature of the circular neighborhood of the stable photovoltaic inspection feature point, and the 4*4 rectangular area corresponding to the stable photovoltaic inspection feature point is obtained. The sum of the horizontal direction value, the sum of the vertical direction value, the sum of the horizontal direction absolute value and the sum of the vertical direction absolute value of the stable photovoltaic inspection feature point are counted to obtain the potential photovoltaic hot spot feature, and the potential photovoltaic hot spot information corresponding to the photovoltaic inspection image is determined according to the potential photovoltaic hot spot feature.

8. An intelligent inspection system for a smart power plant, applied to an intelligent inspection method for a smart power plant as claimed in any one of claims 1 to 7, the inspection system comprising: A model building module, used to collect real-time device data of target devices in a target smart power station, and build a three-dimensional virtual power station model corresponding to the target smart power station according to the real-time device data; A fault location module is used to obtain historical operation data and real-time operation data of the target device, and to construct a device fault prediction model corresponding to the target device based on a machine learning algorithm to identify and locate the device fault location of the target device; A hot spot identification module is used to collect photovoltaic inspection images of photovoltaic modules in the target smart power station through unmanned inspection equipment, and perform hot spot identification on the photovoltaic inspection images to generate potential photovoltaic hot spot information; A surface anomaly module, used to identify the surface abnormality state of the target device based on visual technology and deep learning algorithm; A temperature comparison module is used to build a fire detection network to monitor the current regional temperature of the target smart power station and compare it with a preset temperature threshold to generate corresponding abnormal temperature information; The result sending module is used to generate the intelligent inspection result of the target smart power station and send it to the management end based on the equipment fault location, potential photovoltaic hot spot information, surface abnormal state and abnormal temperature information, combined with the three-dimensional virtual power station model.

9. An electronic device, characterized in that: It includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the intelligent inspection method of the smart power plant as described in any one of claims 1-7.

Images