Operation and maintenance method applied to photovoltaic power station, electronic equipment and readable storage medium
Through the combination of drone inspection and deep learning fault detection models, the problems of high operation and maintenance costs and low efficiency caused by traditional manual inspection are solved, and efficient and accurate photovoltaic power station operation and maintenance management are achieved, reducing operation and maintenance Cost and troubleshooting time.
Patent Information
- Application Number
- CN202410910910.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-08
- Publication Date
- 2024-11-01
AI Technical Summary
The operation and maintenance management of traditional photovoltaic power stations relies on manual inspection, which leads to high operation and maintenance costs and low efficiency, and is difficult to detect and deal with faults in a timely manner, affecting power generation efficiency.
The fault detection model constructed by drones is automatically analyzed by the photovoltaic panel image, identified fault defects, and generated operation and maintenance management reports through the drone control module, information acquisition module, fault detection module and The information generation module realizes automated operation and maintenance.
It significantly improves the operation and maintenance inspection efficiency of photovoltaic power stations, reduces operation and maintenance costs, improves the accuracy and efficiency of fault detection, reduces labor and time costs, and promptly detects and repairs faults to avoid loss of power generation.
Smart Images

Figure CN118884994A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of photovoltaic power generation, and in particular to an operation and maintenance method, electronic equipment and readable storage medium applied to a photovoltaic power station. Background Art
[0002] The development of clean and renewable energy has become a consensus among countries around the world. As an inexhaustible green energy, solar energy has broad application prospects. Solar power generation is the process of converting solar energy into electrical energy using the photovoltaic effect or photothermal effect. Compared with traditional fossil fuel power generation, it has the advantages of being clean, safe, and renewable.
[0003] As the photovoltaic power generation industry develops rapidly, the operation and maintenance of photovoltaic power stations has become a crucial link. As the scale of photovoltaic power stations continues to expand, how to ensure the efficient, stable and safe operation of photovoltaic power stations has become the focus of attention within and outside the industry.
[0004] But at present, the operation and maintenance management of traditional photovoltaic power stations mainly rely on regular manual inspections and maintenance. Since photovoltaic power stations are generally built in remote areas with harsh environments, manual inspections usually require a lot of time and manpower costs, which not only increases the operation and maintenance costs and difficulty of photovoltaic power stations, but also is prone to problems such as untimely detection and low operation and maintenance efficiency. Summary of the invention
[0005] The present application provides an operation and maintenance method, electronic device and readable storage medium applied to a photovoltaic power station, so as to solve the technical problem that the existing photovoltaic power stations adopt manual inspection and operation and maintenance, resulting in high operation and maintenance costs and low efficiency of the photovoltaic power stations.
[0006] According to a first aspect disclosed in the present application, the present application provides an operation and maintenance method applied to a photovoltaic power station, comprising:
[0007] Controlling the drone to inspect the photovoltaic power station based on a preset inspection route, and adjusting the flight attitude of the drone during the inspection process; wherein the flight attitude includes a pitch angle and a yaw angle;
[0008] Acquire the photovoltaic panel image collected by the drone inspection and the position information of the photovoltaic panel corresponding to the photovoltaic panel image;
[0009] Inputting the photovoltaic panel image into a fault detection model for detection to obtain a fault detection result; wherein the fault detection result includes a fault type;
[0010] An operation and maintenance management report is generated based on the fault detection results and location information of the photovoltaic panel.
[0011] In a feasible implementation manner, the adjusting the flight attitude of the UAV includes:
[0012] Acquire flight parameters of the drone; wherein the flight parameters include the relative height and relative distance between the drone and the photovoltaic panel;
[0013] Based on the relative height, adjusting the pitch angle of the drone;
[0014] Based on the relative distance, the yaw angle of the UAV is adjusted.
[0015] In a feasible implementation manner, adjusting the pitch angle of the drone based on the relative height includes:
[0016] Comparing the relative height with a preset first height threshold and a second height threshold; wherein the first height threshold is less than the second height threshold;
[0017] If the relative height is less than the first height threshold, increasing the pitch angle;
[0018] If the relative height is greater than the second height threshold, reducing the pitch angle;
[0019] If the relative height is greater than the first height threshold and less than the second height threshold, the pitch angle is kept unchanged.
[0020] In a feasible implementation manner, the method further includes:
[0021] Determining a trigger strength coefficient based on the flight parameters and a preset membership function;
[0022] Obtaining a first adjustment amplitude of the pitch angle based on a product of the trigger intensity coefficient and a preset first amplitude threshold;
[0023] The pitch angle is adjusted based on the first adjustment amplitude.
[0024] In a feasible implementation manner, adjusting the yaw angle of the drone based on the relative distance includes:
[0025] Comparing the relative distance with a preset first distance threshold and a second distance threshold; wherein the first distance threshold is less than the second distance threshold;
[0026] If the relative distance is less than the first distance threshold, adjusting the yaw angle to move the drone away from the photovoltaic panel;
[0027] If the relative distance is greater than the second distance threshold, adjusting the yaw angle to make the drone approach the photovoltaic panel;
[0028] If the relative distance is greater than the first distance threshold and less than the second distance threshold, the yaw angle is kept unchanged.
[0029] In a feasible implementation manner, the method further includes:
[0030] Determining a trigger strength coefficient based on the flight parameter input and a preset membership function;
[0031] Obtaining a second adjustment amplitude of the yaw angle based on a product of the trigger intensity coefficient and a preset second amplitude threshold;
[0032] The yaw angle is adjusted based on the second adjustment amplitude.
[0033] In a feasible implementation manner, the flight parameter further includes a flight speed, and determining the trigger intensity coefficient based on the flight parameter and a preset membership function includes:
[0034] Inputting the relative height into a preset height membership function for solving, and obtaining a height membership degree;
[0035] Inputting the relative distance into a preset distance membership function for solving, and obtaining the distance membership degree;
[0036] Inputting the flight speed into a preset speed membership function for solving, and obtaining a speed membership degree;
[0037] The minimum value among the height membership, the distance membership and the speed membership is selected to obtain a trigger strength coefficient.
[0038] In a feasible implementation, the fault detection model is a composite convolutional neural network model, which includes an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a fully connected layer and an output layer connected in sequence, and the method further includes:
[0039] Obtaining a training set and a test set; wherein the training set and the test set are marked with real labels;
[0040] Inputting the training set into the composite convolutional neural model to obtain a training result;
[0041] Based on the training result and a preset loss function, obtaining a loss amount between the training result and the true label;
[0042] The composite convolutional neural network model is tested based on the test set. When the test result of the test set converges, the training is terminated; otherwise, the network parameters of the composite convolutional neural network model are optimized according to the loss amount, and the training set is re-input into the composite convolutional neural model for training until the test result converges to obtain the fault detection model.
[0043] According to a second aspect disclosed in the present application, the present application provides an operation and maintenance device applied to a photovoltaic power station, comprising:
[0044] A drone control module is used to control the drone to inspect the photovoltaic power station based on a preset inspection route, and to adjust the flight attitude of the drone during the drone inspection process; wherein the flight attitude includes a pitch angle and a yaw angle;
[0045] An information collection module, used to obtain the photovoltaic panel image collected by the drone inspection, and the position information of the photovoltaic panel corresponding to the photovoltaic panel image;
[0046] A fault detection module, used for inputting the photovoltaic panel image into a fault detection model for detection to obtain a fault detection result; wherein the fault detection result includes a fault type;
[0047] The information generation module is used to generate an operation and maintenance management report based on the fault detection result and location information of the photovoltaic panel.
[0048] According to a third aspect disclosed in the present application, there is provided an electronic device, comprising a processor, and a memory communicatively connected to the processor;
[0049] The memory stores computer-executable instructions;
[0050] The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of the first aspects.
[0051] According to a fourth aspect disclosed in the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement any one of the methods described in the first aspect.
[0052] According to a fifth aspect disclosed in the present application, a computer program product is provided, including a computer program, wherein the computer program is used to implement any one of the methods in the first aspect when executed by a processor.
[0053] Compared with the prior art, this application has the following beneficial effects:
[0054] The present application provides an operation and maintenance method, electronic device and readable storage medium for photovoltaic power stations. The photovoltaic power station is inspected by drones. The drones can quickly and efficiently cover a large area of the photovoltaic power station, significantly improving the efficiency of the operation and maintenance inspection of the photovoltaic power station and reducing the operation and maintenance cost of the photovoltaic power station. Compared with the traditional manual inspection operation and maintenance mode, drones can complete the operation and maintenance inspection of a large number of photovoltaic panels in a photovoltaic power station in a relatively short time. At the same time, the fault detection model constructed using the deep learning algorithm can automatically analyze the photovoltaic panel images collected by the drone, accurately identify the fault defects of the photovoltaic panel, greatly improve the accuracy and efficiency of photovoltaic panel detection, and further reduce the operation and maintenance cost of the photovoltaic power station. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0056] Figure 1 A schematic diagram of a flow chart of an operation and maintenance method applied to a photovoltaic power station provided in an embodiment of the present application;
[0057] Figure 2 A flow chart of another operation and maintenance method for a photovoltaic power station provided in an embodiment of the present application;
[0058] Figure 3 A schematic diagram of the structure of an operation and maintenance device applied to a photovoltaic power station provided in an embodiment of the present application;
[0059] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.
[0060] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION
[0061] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0062] The development of clean and renewable energy has become a consensus among countries around the world. As an inexhaustible green energy, solar energy has broad application prospects. Solar power generation is the process of converting solar energy into electrical energy using the photovoltaic effect or photothermal effect. Compared with traditional fossil fuel power generation, it has the advantages of being clean, safe, and renewable.
[0063] As the photovoltaic power generation industry develops rapidly, the operation and maintenance of photovoltaic power stations has become a crucial link. As the scale of photovoltaic power stations continues to expand, how to ensure the efficient, stable and safe operation of photovoltaic power stations has become the focus of attention within and outside the industry.
[0064] However, at present, the operation and maintenance of traditional photovoltaic power stations mainly rely on manual regular inspection and maintenance. Since photovoltaic power stations are generally built in remote areas with harsh environments, manual inspection usually requires a lot of time and manpower costs, which not only increases the operation and maintenance costs and difficulties of photovoltaic power stations, but also easily leads to problems such as untimely detection and low operation and maintenance efficiency. In addition, since manual regular maintenance also has lags, it is impossible to detect and deal with the failure of photovoltaic panels in photovoltaic power stations in a timely manner, which easily leads to a reduction in the power generation efficiency of photovoltaic power stations and causes economic losses.
[0065] In response to the above technical problems, this application proposes an operation and maintenance method applied to photovoltaic power stations. Through the photovoltaic power station operation and maintenance mode that combines drone inspection with intelligent detection of photovoltaic panels, the operation and maintenance efficiency of photovoltaic power stations is effectively improved and the operation and maintenance costs of photovoltaic power stations are reduced.
[0066] The following is a detailed description of the technical solution of the operation and maintenance method for photovoltaic power stations provided by the present application through specific embodiments. It should be noted that the following embodiments may exist alone or in combination with each other, and the same or similar contents may not be repeated in different embodiments.
[0067] Figure 1 A schematic diagram of a process flow of an operation and maintenance method for a photovoltaic power station provided in an embodiment of the present application is provided in Figure 1 In some embodiments, the process of the operation and maintenance method applied to a photovoltaic power station includes the following steps:
[0068] S101, controlling the drone to inspect the photovoltaic power station based on a preset inspection route, and adjusting the flight attitude of the drone during the drone inspection process; wherein the flight attitude includes a pitch angle and a yaw angle.
[0069] Among them, the drone is controlled to inspect the photovoltaic power station along the preset inspection route, and the flight posture of the drone is continuously adjusted during the inspection process to ensure that the drone can effectively photograph the photovoltaic panels in the photovoltaic power station.
[0070] Specifically, before the drone inspects the photovoltaic power station, the inspection route of the drone can be planned based on the distribution of photovoltaic panels in the photovoltaic power station to further improve the inspection efficiency of the photovoltaic panels in the photovoltaic power station.
[0071] S102, obtaining the photovoltaic panel image collected by the drone inspection and the position information of the corresponding photovoltaic panel in the photovoltaic panel image.
[0072] The photovoltaic panel images taken by the drone during the inspection process and the location information of the photovoltaic panels photographed in the photovoltaic panel images are obtained. Specifically, the location information can be marked with longitude and latitude, or with the relative position of the photovoltaic panels in the panoramic view of the photovoltaic power station.
[0073] S103, inputting the photovoltaic panel image into a fault detection model for detection to obtain a fault detection result; wherein the fault detection result includes a fault type.
[0074] The photovoltaic panel image is input into the fault detection model, and the fault detection model is used to automatically complete the fault detection of the photovoltaic panel and output the fault detection result. When a fault is detected in the photovoltaic panel image, the fault detection model can also output the fault type to further facilitate the operation and maintenance personnel of the photovoltaic power station to know the fault status of the photovoltaic panel.
[0075] Specifically, the failure types of photovoltaic panels include hot spots, fragmentation, shading, burning, cracks, etc.
[0076] S104, generating an operation and maintenance management report based on the fault detection results and location information of the photovoltaic panels.
[0077] Among them, an operation and maintenance management report of the photovoltaic power station is generated according to the fault detection results and location information of the photovoltaic panels in the photovoltaic power station. The operation and maintenance management report can provide data support for the operation and maintenance management of the photovoltaic power station, help operation and maintenance personnel to formulate reasonable maintenance plans, improve operation and maintenance efficiency, and reduce operation and maintenance costs.
[0078] Specifically, the operation and maintenance management report includes panoramic images of the photovoltaic power station and defective photovoltaic panels, as well as detailed information such as the fault type, fault location, and location information of the defective photovoltaic panels.
[0079] In this embodiment, the photovoltaic power station is inspected by drones. The drones can quickly and efficiently cover a large area of the photovoltaic power station, significantly improving the efficiency of the operation and maintenance inspection of the photovoltaic power station and reducing the operation and maintenance cost of the photovoltaic power station. Compared with the traditional manual inspection operation and maintenance mode, drones can complete the operation and maintenance inspection of a large number of photovoltaic panels in a photovoltaic power station in a relatively short time. At the same time, the fault detection model constructed using the deep learning algorithm can automatically analyze the photovoltaic panel images collected by the drone and accurately identify the fault defects of the photovoltaic panel, greatly improving the accuracy and efficiency of photovoltaic panel detection, and further reducing the operation and maintenance cost of the photovoltaic power station.
[0080] In summary, the photovoltaic power station operation and maintenance mode that combines drone inspections with intelligent detection of photovoltaic panels reduces the manpower and time costs of manual inspections through automated and intelligent detection methods, and can promptly detect and repair photovoltaic panel failures, avoiding power generation losses caused by failures and reducing the operation and maintenance costs of photovoltaic power stations.
[0081] exist Figure 1 Based on the embodiment shown, the following Figure 2 , the technical solution of the above-mentioned operation and maintenance method applied to photovoltaic power stations is further introduced.
[0082] Figure 2 Another flow chart of an operation and maintenance method for a photovoltaic power station provided in an embodiment of the present application is shown in FIG. Figure 2 In some embodiments, the process of the operation and maintenance method applied to a photovoltaic power station includes the following steps:
[0083] S201, controlling the drone to inspect the photovoltaic power station based on a preset inspection route.
[0084] It should be noted that the execution process of step S201 is the same as the execution process of step S101, which will not be repeated here.
[0085] S202, during the drone inspection process, obtaining the flight parameters of the drone; wherein the flight parameters include the relative height and relative distance between the drone and the photovoltaic panel.
[0086] Among them, during the drone inspection process, the drone's flight parameters can reflect the positional relationship between the drone and the photovoltaic panel, so as to adjust the relative position between the drone and the photovoltaic panel according to the flight parameters to ensure that the drone can effectively photograph the photovoltaic panel under safe conditions.
[0087] S203, comparing the relative height with a preset first height threshold and a second height threshold; wherein the first height threshold is smaller than the second height threshold.
[0088] The relative height between the drone and the photovoltaic panel is compared with a preset height threshold to determine the height relationship between the drone and the photovoltaic panel, so as to determine whether the flight altitude of the drone needs to be adjusted. This is because if the height between the drone and the photovoltaic panel is too large, the drone may not be able to clearly photograph the photovoltaic panel, and if the height between the drone and the photovoltaic panel is too small, the drone is in danger of colliding with the photovoltaic panel.
[0089] S204: If the relative height is less than the first height threshold, increase the pitch angle.
[0090] Among them, when the relative height is less than the first height threshold, it indicates that the relative height between the drone and the photovoltaic panel is too small, and the relative height between the drone and the photovoltaic panel is increased by increasing the pitch angle of the drone.
[0091] S205: If the relative height is greater than a second height threshold, reduce the pitch angle.
[0092] Among them, when the relative height is greater than the second height threshold, it indicates that the relative height between the drone and the photovoltaic panel is too large, and the relative height between the drone and the photovoltaic panel is reduced by reducing the pitch angle of the drone.
[0093] S206: If the relative height is greater than the first height threshold and less than the second height threshold, keep the pitch angle unchanged.
[0094] Among them, when the relative height is greater than the first height threshold and less than the second height threshold, it indicates that the relative height between the UAV and the photovoltaic panel is appropriate, and there is no need to adjust the pitch angle of the UAV, and the current pitch angle remains unchanged.
[0095] Preferably, during the pitch angle adjustment process, the pitch angle adjustment amplitude also needs to be determined, so as to complete the pitch angle adjustment through the adjustment amplitude and avoid the situation where the adjustment amplitude is too large and causes the drone to lose control.
[0096] Step 1: Determine the trigger strength coefficient based on the flight parameters and the preset membership function.
[0097] Specifically, the pitch angle adjustment range is determined by the relative height, relative distance and flight speed. By calculating these three factors and the membership function, the trigger intensity coefficient that determines the adjustment range can be obtained. The specific process of obtaining the trigger intensity coefficient includes:
[0098] Step 101: input the relative height into a preset height membership function for solution to obtain the height membership degree.
[0099] Specifically, the height membership function satisfies the following formula:
[0100]
[0101] Among them, μ H_low Represents the first degree of membership, μ H_high represents the second height membership, H represents the relative height, H low Represents the first height threshold, H high Indicates the second height threshold.
[0102] For example, when H≤H low When H_low (H) = 1, μ H_high (H) = 0, which means that the current flight altitude completely belongs to the Low fuzzy set, and the altitude membership value is μ H_low (H) = 1. When H low <H<H high When 0<μ H_low (H)<1,0<μ H_high (H)<1, at this time, it means that the current flight altitude belongs to both the Low fuzzy set and the High fuzzy set, and the altitude membership value is μ H_low (H), μ H_high (H). When H≥H high When H_low (H) = 0, μ H_high (H) = 1, which means that the current flight altitude completely belongs to the High fuzzy set, and the altitude membership value is μ H_high (H) = 1.
[0103] Specifically, the following distance membership and speed membership are calculated according to the above value-taking rules.
[0104] Step 102: input the relative distance into a preset distance membership function for solution to obtain the distance membership degree.
[0105] Specifically, the distance membership function satisfies the following formula:
[0106]
[0107] Among them, μ D_near Represents the first distance membership, μ D_far represents the second distance membership, D represents the relative distance, D near represents the first distance threshold, D far Indicates the second distance threshold.
[0108] Step 103, input the flight speed into a preset speed membership function for solution to obtain the speed membership.
[0109] Specifically, the speed membership function satisfies the following formula:
[0110]
[0111] Among them, μ V_slow represents the first velocity membership, μ V_fast represents the second speed membership, V represents the flight speed, V slow Represents the first speed threshold, V fast Indicates the second speed threshold.
[0112] Step 104, selecting the minimum value among the height membership, the distance membership and the speed membership to obtain the trigger strength coefficient.
[0113] Among them, after calculating the height membership, distance membership and speed membership, the minimum value among the three is obtained as the trigger strength coefficient.
[0114] Specifically, the trigger strength coefficient satisfies the following formula:
[0115] α=min(μ H ,μ D ,μ V )
[0116] Among them, α represents the trigger strength coefficient, min represents the minimum value, μ H Represents the high degree of membership, μ D Represents the distance membership, μ V Indicates speed membership.
[0117] Specifically, according to the value range of the membership degree, the trigger strength coefficient α∈(0,1].
[0118] Step 2: Obtain a first adjustment amplitude of the pitch angle based on the product of the trigger intensity coefficient and a preset first amplitude threshold.
[0119] Among them, the first amplitude threshold is the preset maximum adjustable amplitude of the pitch angle. By triggering the intensity coefficient, the pitch angle adjustment amplitude can be dynamically adjusted based on the maximum adjustable amplitude to prevent the drone from losing control during the pitch angle adjustment process.
[0120] Specifically, the first adjustment range satisfies the following formula:
[0121] Δ pitch =α*A1
[0122] Among them, Δ pitch represents the first adjustment amplitude, α represents the trigger strength coefficient, and A1 represents the first amplitude threshold.
[0123] Step 3: Adjust the pitch angle based on the first adjustment amplitude.
[0124] After the first adjustment range of the pitch angle is determined, the pitch angle is increased or decreased according to the first adjustment range.
[0125] S207, comparing the relative distance with a preset first distance threshold and a second distance threshold; wherein the first distance threshold is smaller than the second distance threshold.
[0126] The relative distance between the drone and the photovoltaic panel is compared with the preset distance threshold to determine the distance relationship between the drone and the photovoltaic panel to determine whether the flight direction of the drone needs to be adjusted. This is because if the distance between the drone and the photovoltaic panel is too large, the drone may not be able to clearly photograph the photovoltaic panel, and if the distance between the drone and the photovoltaic panel is too small, the drone is in danger of colliding with the photovoltaic panel.
[0127] S208, if the relative distance is less than the first distance threshold, adjusting the yaw angle to keep the drone away from the photovoltaic panel;
[0128] When the relative distance is less than the first distance threshold, it indicates that the relative distance between the UAV and the photovoltaic panel is too small, and the UAV is moved away from the photovoltaic panel by changing the yaw angle of the UAV.
[0129] S209: If the relative distance is greater than the second distance threshold, adjust the yaw angle to bring the drone closer to the photovoltaic panel.
[0130] When the relative distance is greater than the second distance threshold, it indicates that the relative distance between the UAV and the photovoltaic panel is too large, and the UAV is brought closer to the photovoltaic panel by changing the yaw angle of the UAV.
[0131] S210: If the relative distance is greater than the first distance threshold and less than the second distance threshold, the yaw angle is kept unchanged.
[0132] Among them, when the relative distance is greater than the first distance threshold and less than the second distance threshold, it indicates that the relative distance between the UAV and the photovoltaic panel is appropriate, and there is no need to adjust the pitch angle of the UAV, and the current pitch angle remains unchanged.
[0133] Preferably, during the adjustment of the yaw angle, the adjustment amplitude of the yaw angle also needs to be obtained, so as to complete the adjustment of the yaw angle through the adjustment amplitude and avoid the situation where the adjustment amplitude is too large and causes the drone to lose control.
[0134] Step 1: Determine the trigger intensity coefficient based on the flight parameter input and the preset membership function.
[0135] Step 2: Obtain a second adjustment amplitude of the yaw angle based on the product of the trigger intensity coefficient and a preset second amplitude threshold.
[0136] Among them, the second amplitude threshold is the preset maximum adjustable amplitude of the yaw angle. By triggering the intensity coefficient, the size of the yaw angle adjustment amplitude can be dynamically adjusted based on the maximum adjustable amplitude to prevent the drone from losing control during the yaw angle adjustment process.
[0137] Specifically, the first adjustment range satisfies the following formula:
[0138] Δ yaw =α*A2
[0139] Among them, Δ yaw represents the second adjustment amplitude, α represents the trigger strength coefficient, and A2 represents the second amplitude threshold.
[0140] Step 3: Adjust the yaw angle based on the second adjustment amplitude.
[0141] S211, obtaining the photovoltaic panel image collected by the drone inspection and the position information of the corresponding photovoltaic panel in the photovoltaic panel image.
[0142] It should be noted that the execution process of step S211 is the same as the execution process of step S103, which will not be repeated here.
[0143] S212, inputting the photovoltaic panel image into a fault detection model for detection to obtain a fault detection result; wherein the fault detection result includes a fault type.
[0144] Preferably, the fault detection model is a composite convolutional neural network model, which includes an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a fully connected layer and an output layer connected in sequence.
[0145] Specifically, the input layer is used to input the original data, i.e., the photovoltaic panel image, into the fault detection model.
[0146] Specifically, the first convolutional layer is used to perform the first feature extraction on the input data. The local features of the image are extracted through the convolution operation of the convolution kernel (also called filter or neuron) and the input data, and the nonlinear activation function. Among them, the first convolutional layer satisfies the following formula:
[0147] Conv1(I)=ReLU(W1*I+b1)
[0148] Among them, Conv1 represents the first convolution operation, I represents the input image, ReLU represents the activation function, W1 represents the convolution kernel of the convolution layer, and b1 represents the bias term of the first convolution layer.
[0149] Specifically, the first pooling layer is used to downsample the feature map output by the first convolutional layer, reduce the dimension of the feature map, and retain important features, thereby reducing the number of parameters and enhancing the robustness of the features. This helps prevent overfitting and speed up the calculation. The first pooling layer satisfies the following formula:
[0150] Pool1(Conv1)=max(Conv1)
[0151] Among them, Pool1 represents the first pooling operation, and max represents the maximum value function.
[0152] Specifically, the second convolutional layer is used to perform a second feature extraction on the output of the first pooling layer, further abstracting and combining low-level features to form a higher-level feature representation. The second convolutional layer satisfies the following formula:
[0153] Conv2(Pool1)=ReLU(W2*Pool1+b2)
[0154] Among them, Conv2 represents the second convolution operation, ReLU represents the activation function, W2 represents the convolution kernel of the second convolution layer, and b4 represents the bias term of the second convolution layer.
[0155] Specifically, the second pooling layer is similar to the first pooling layer, and downsamples the output of the second convolutional layer to further reduce the dimension of the feature map and enhance the robustness of the feature. The second pooling layer satisfies the following formula:
[0156] Pool2(Conv2)=max(Conv2)
[0157] Among them, Pool2 represents the second pooling operation, and max represents the maximum value function.
[0158] Specifically, the fully connected layer is used to map the features extracted by the previous convolutional layer and pooling layer to the sample label space, that is, to convert the feature representation into a specific category or label. Among them, the fully connected layer satisfies the following formula:
[0159] FC(Pool2)=ReLU(W3*Pool2+b3)
[0160] Among them, FC represents the fully connected operation, ReLU represents the activation function, W3 represents the weight matrix of the fully connected layer, and b3 represents the bias term of the fully connected layer.
[0161] Specifically, the output layer is used to output the network's prediction results for the input data. For classification tasks, the output layer usually contains a Softmax function to convert the output of the fully connected layer into a probability distribution. The output layer satisfies the following formula:
[0162] Output(FC)=Softmax(W4*FC+b4)
[0163] Among them, Output represents the classification output operation, Softmax represents the activation function, W4 represents the weight matrix of the output layer, and b4 represents the bias term of the output layer.
[0164] Specifically, under the structure of the composite convolutional neural network, the process of training the composite convolutional neural network as a fault detection model includes:
[0165] Step 1: Obtain a training set and a test set; wherein the training set and the test set are annotated with true labels.
[0166] The training set is a dataset used for model training, while the test set is a dataset used to evaluate model performance. Specifically, both the training set and the test set contain multiple photovoltaic panel images, and the photovoltaic panel images are manually annotated with real labels, which reflect information such as the fault type and fault location of the photovoltaic panel images.
[0167] Step 2: Input the training set into the composite convolutional neural model to obtain the training results.
[0168] Step 3: Based on the training results and the preset loss function, obtain the loss between the training results and the true label.
[0169] Among them, the loss function satisfies the following formula:
[0170]
[0171] Among them, J represents the objective function, θ represents the parameters of the composite convolutional neural network model, m represents the number of samples, and y i represents the label of the i-th sample, I i represents the input image of the i-th sample, n represents the neurons in the output layer, W4 (j) represents the weight of the jth neuron in the output layer, b4 (j) Represents the bias of the jth neuron in the output layer.
[0172] Among them, FC represents the fully connected operation, Pool2 represents the second pooling operation, Conv2 represents the second convolution operation, Pool1 represents the first pooling operation, and Conv1 represents the first convolution operation.
[0173] Step 4: Test the composite convolutional neural network model based on the test set. When the test results of the test set converge, the training is terminated; otherwise, the network parameters of the composite convolutional neural network model are optimized according to the loss amount, and the training set is re-input into the composite convolutional neural model for training until the test results converge to obtain the fault detection model.
[0174] Among them, the composite convolutional neural network is iteratively trained and tested to meet the detection requirements of photovoltaic panel faults and obtain a fault detection model.
[0175] S213, generating an operation and maintenance management report based on the fault detection results and location information of the photovoltaic panels.
[0176] It should be noted that the execution process of step S213 is the same as the execution process of step S104, which will not be repeated here.
[0177] In this embodiment, the photovoltaic power station operation and maintenance mode that combines drone inspection with photovoltaic panel intelligent detection effectively improves the operation and maintenance efficiency of the photovoltaic power station and reduces the operation and maintenance cost of the photovoltaic power station.
[0178] Figure 3 is a schematic diagram of a structure of an operation and maintenance device for a photovoltaic power station provided in an embodiment of the present application, see Figure 3 The operation and maintenance device applied to a photovoltaic power station includes various functional modules for implementing the aforementioned operation and maintenance method applied to a photovoltaic power station, and any functional module can be implemented by software and / or hardware.
[0179] In some embodiments, the operation and maintenance device 300 applied to a photovoltaic power station includes a drone control module 301, an information collection module 302, a fault detection module 303 and an information generation module 304. Among them:
[0180] The drone control module 301 is used to control the drone to inspect the photovoltaic power station based on a preset inspection route, and adjust the flight attitude of the drone during the drone inspection process; wherein the flight attitude includes a pitch angle and a yaw angle;
[0181] The information collection module 302 is used to obtain the photovoltaic panel images collected by the drone inspection and the location information of the corresponding photovoltaic panel in the photovoltaic panel image;
[0182] The fault detection module 303 is used to input the photovoltaic panel image into the fault detection model for detection to obtain a fault detection result; wherein the fault detection result includes the fault type;
[0183] The information generation module 304 is used to generate an operation and maintenance management report based on the fault detection results and location information of the photovoltaic panels.
[0184] In some embodiments, the drone control module 301 is specifically used to:
[0185] Obtaining the flight parameters of the UAV, wherein the flight parameters include the relative height and relative distance between the UAV and the photovoltaic panel;
[0186] Adjust the pitch angle of the drone based on the relative altitude;
[0187] Based on the relative distance, the yaw angle of the drone is adjusted.
[0188] In some embodiments, the drone control module 301 is further used to:
[0189] Comparing the relative height with a preset first height threshold and a second height threshold; wherein the first height threshold is less than the second height threshold;
[0190] If the relative height is less than the first height threshold, increase the pitch angle;
[0191] If the relative height is greater than a second height threshold, reducing the pitch angle;
[0192] If the relative altitude is greater than the first altitude threshold and less than the second altitude threshold, the pitch angle is kept unchanged.
[0193] In some embodiments, the drone control module 301 is further used to:
[0194] Determine the trigger strength coefficient based on the flight parameters and the preset membership function;
[0195] Obtaining a first adjustment amplitude of the pitch angle based on a product of the trigger intensity coefficient and a preset first amplitude threshold;
[0196] The pitch angle is adjusted based on the first adjustment amplitude.
[0197] In some embodiments, the drone control module 301 is further used to:
[0198] Compare the relative distance with a preset first distance threshold and a second distance threshold; wherein the first distance threshold is smaller than the second distance threshold;
[0199] If the relative distance is less than the first distance threshold, the yaw angle is adjusted to keep the drone away from the photovoltaic panel;
[0200] If the relative distance is greater than a second distance threshold, the yaw angle is adjusted to bring the drone closer to the photovoltaic panel;
[0201] If the relative distance is greater than the first distance threshold and less than the second distance threshold, the yaw angle is kept unchanged.
[0202] In some embodiments, the drone control module 301 is further used to:
[0203] Determine the trigger strength coefficient based on the flight parameter input and the preset membership function;
[0204] Obtaining a second adjustment amplitude of the yaw angle based on a product of the trigger intensity coefficient and a preset second amplitude threshold;
[0205] The yaw angle is adjusted based on the second adjustment amplitude.
[0206] In some embodiments, the drone control module 301 is further used to:
[0207] Input the relative height into the preset height membership function to solve and obtain the height membership degree;
[0208] Input the relative distance into the preset distance membership function for solution to obtain the distance membership degree;
[0209] Input the flight speed into the preset speed membership function to solve and obtain the speed membership;
[0210] The minimum value among the height membership, distance membership and speed membership is selected to obtain the trigger strength coefficient.
[0211] In some embodiments, the fault detection model is a composite convolutional neural network model, which includes an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a fully connected layer and an output layer connected in sequence. The fault detection module 303 is also specifically used to:
[0212] Get the training set and the test set, where the training set and the test set are marked with real labels;
[0213] Input the training set into the composite convolutional neural model to obtain the training results;
[0214] Based on the training results and the preset loss function, obtain the loss between the training results and the true label;
[0215] The composite convolutional neural network model is tested based on the test set. When the test results of the test set converge, the training is terminated; otherwise, the network parameters of the composite convolutional neural network model are optimized according to the loss amount, and the training set is re-input into the composite convolutional neural model for training until the test results converge and the fault detection model is obtained.
[0216] The operation and maintenance device 300 applied to a photovoltaic power station provided in the embodiment of the present application is used to execute the technical solution provided in the aforementioned operation and maintenance method embodiment applied to a photovoltaic power station. Its implementation principle and technical effect are similar to those in the aforementioned method embodiment and will not be repeated here.
[0217] It should be noted that it should be understood that the division of the various modules of the above device is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software called by processing elements, or all in the form of hardware, or some modules can be implemented in the form of software called by processing elements, and some modules can be implemented in the form of hardware. For example, the drone control module 301 can be a separately established processing element, or it can be integrated in a chip of the above device. In addition, it can also be stored in the memory of the above device in the form of program code, and called and executed by a processing element of the above device. The function of the drone control module 301. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or instructions in software form.
[0218] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application, see Figure 4 , the electronic device 400 includes: a processor 401, and a memory 402 communicatively connected to the processor 401;
[0219] Memory 402 stores computer executable instructions;
[0220] The processor 401 executes the computer-executable instructions stored in the memory 402 to implement the above-mentioned technical solution of the operation and maintenance method applied to the photovoltaic power station.
[0221] In the above-mentioned electronic device 400, the memory 402 and the processor 401 are electrically connected directly or indirectly to realize data transmission or interaction. For example, these elements can be electrically connected to each other through one or more communication buses or signal lines, such as through a bus connection. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc., but it does not mean that there is only one bus or one type of bus. The memory 402 stores computer execution instructions for implementing the aforementioned operation and maintenance method applied to a photovoltaic power station, including at least one software function module that can be stored in the memory 402 in the form of software or firmware. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 402.
[0222] The memory 402 includes at least one type of readable storage medium, including but not limited to random access memory (RAM), read only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), and electrically erasable read-only memory (EEPROM). The memory 402 is used to store programs, and the processor 401 executes the programs after receiving the execution instruction. Furthermore, the software programs and modules in the above-mentioned memory 402 may also include an operating system, which may include various software components and / or drivers for managing system tasks (such as memory management, storage device control, power management, etc.), and may communicate with various hardware or software components to provide an operating environment for other software components.
[0223] The processor 401 may be an integrated circuit chip having the ability to process signals. The processor 401 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), etc. The methods, steps, and logic diagrams disclosed in the embodiments of the present application may be implemented or executed. The general-purpose processor may be a microprocessor, or the processor 401 may be any conventional processor, etc.
[0224] The electronic device 400 is used to execute the technical solution provided by the aforementioned operation and maintenance method embodiment applied to a photovoltaic power station. Its implementation principle and technical effect are similar to those in the aforementioned method embodiment and will not be repeated here.
[0225] The embodiment of the present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the technical solution of the operation and maintenance method applied to a photovoltaic power station as described above is implemented.
[0226] The computer-readable storage medium may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The computer-readable storage medium may be any available medium that can be accessed by a general or special purpose computer.
[0227] An exemplary 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). Of course, the processor and the readable storage medium can also exist as discrete components in a control device for an operation and maintenance device applied to a photovoltaic power station.
[0228] The embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, is used to implement the technical solution of the operation and maintenance method applied to a photovoltaic power station as described above.
[0229] In the above embodiments, it will be appreciated by those skilled in the art that the above-mentioned various method embodiments can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loading and executing computer program instructions on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless network, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or a data center that contains one or more available media integrations. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).
[0230] In the above embodiments, the description of each embodiment has its own emphasis. For the part not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0231] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.
[0232] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.
Claims
1. An operation and maintenance method applied to a photovoltaic power station, characterized in that: include: Controlling the drone to inspect the photovoltaic power station based on a preset inspection route, and adjusting the flight attitude of the drone during the inspection process; wherein the flight attitude includes a pitch angle and a yaw angle; Acquire the photovoltaic panel image collected by the drone inspection and the position information of the photovoltaic panel corresponding to the photovoltaic panel image; Inputting the photovoltaic panel image into a fault detection model for detection to obtain a fault detection result; wherein the fault detection result includes a fault type; An operation and maintenance management report is generated based on the fault detection results and location information of the photovoltaic panel.
2. The method according to claim 1, characterized in that: The step of adjusting the flight attitude of the UAV includes: Acquire flight parameters of the drone; wherein the flight parameters include the relative height and relative distance between the drone and the photovoltaic panel; Based on the relative height, adjusting the pitch angle of the drone; Based on the relative distance, the yaw angle of the UAV is adjusted.
3. The method according to claim 2, characterized in that The adjusting the pitch angle of the UAV based on the relative height includes: Comparing the relative height with a preset first height threshold and a second height threshold; wherein the first height threshold is less than the second height threshold; If the relative height is less than the first height threshold, increasing the pitch angle; If the relative height is greater than the second height threshold, reducing the pitch angle; If the relative height is greater than the first height threshold and less than the second height threshold, the pitch angle is kept unchanged.
4. The method according to claim 3, characterized in that The method further comprises: Determining a trigger strength coefficient based on the flight parameters and a preset membership function; Obtaining a first adjustment amplitude of the pitch angle based on a product of the trigger intensity coefficient and a preset first amplitude threshold; The pitch angle is adjusted based on the first adjustment amplitude.
5. The method according to claim 2, characterized in that: The adjusting the yaw angle of the UAV based on the relative distance includes: Comparing the relative distance with a preset first distance threshold and a second distance threshold; wherein the first distance threshold is less than the second distance threshold; If the relative distance is less than the first distance threshold, adjusting the yaw angle to move the drone away from the photovoltaic panel; If the relative distance is greater than the second distance threshold, adjusting the yaw angle to make the drone approach the photovoltaic panel; If the relative distance is greater than the first distance threshold and less than the second distance threshold, the yaw angle is kept unchanged.
6. The method according to claim 5, characterized in that The method further comprises: Determining a trigger strength coefficient based on the flight parameter input and a preset membership function; Obtaining a second adjustment amplitude of the yaw angle based on a product of the trigger intensity coefficient and a preset second amplitude threshold; The yaw angle is adjusted based on the second adjustment amplitude.
7. The method according to claim 4 or 6, characterized in that: The flight parameters also include a flight speed, and determining the trigger strength coefficient based on the flight parameters and a preset membership function includes: Inputting the relative height into a preset height membership function for solving, and obtaining a height membership degree; Inputting the relative distance into a preset distance membership function for solving, and obtaining the distance membership degree; Inputting the flight speed into a preset speed membership function for solving, and obtaining a speed membership degree; The minimum value among the height membership, the distance membership and the speed membership is selected to obtain a trigger strength coefficient.
8. The method according to claim 1, characterized in that The fault detection model is a composite convolutional neural network model, which includes an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a fully connected layer and an output layer connected in sequence, and the method further includes: Obtaining a training set and a test set; wherein the training set and the test set are marked with real labels; Inputting the training set into the composite convolutional neural model to obtain a training result; Based on the training result and a preset loss function, obtaining a loss amount between the training result and the true label; The composite convolutional neural network model is tested based on the test set. When the test result of the test set converges, the training is terminated; otherwise, the network parameters of the composite convolutional neural network model are optimized according to the loss amount, and the training set is re-input into the composite convolutional neural model for training until the test result converges to obtain the fault detection model.
9. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 8.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 8 when executed by a processor.
