Photovoltaic module cleaning early warning method, device and equipment and storage medium

Abstract

The invention discloses a photovoltaic module cleaning early warning method, device and equipment and a storage medium, and the method comprises the steps: obtaining a current cleaning period according to a cleaning period model; the last cleaning date is obtained, and the current cleaning date is determined according to the current cleaning period and the last cleaning date; obtaining current panel image data of the target photovoltaic module, and inputting the current panel image data into a pre-constructed cleanliness model to output the current cleanliness of the target photovoltaic module; obtaining the current actual power generation efficiency of the target photovoltaic module, and inputting the current actual power generation efficiency into a pre-established specific power generation efficiency model to output the current specific power generation efficiency of the target photovoltaic module; obtaining a cleanliness threshold value and a specific power generation efficiency threshold value according to the current cleaning date, the cleanliness model and the specific power generation efficiency model; and when the current cleanliness of the target photovoltaic module is lower than a cleanliness threshold value and / or the current specific power generation efficiency is lower than a specific power generation efficiency threshold value, triggering cleaning early warning. According to the invention, the timeliness and accuracy of cleaning early warning can be improved.

Description

Technical Field

[0001] The present application relates to the field of photovoltaic cleaning, and in particular to a photovoltaic module cleaning early warning method, device, equipment and storage medium. Background Art

[0002] Photovoltaic power plants, as a crucial component of clean energy, have been widely promoted and applied worldwide in recent years. However, the continuous accumulation of dust on photovoltaic panels can reduce the cleanliness of the panels' surfaces, impacting the power generation efficiency of the power plant and potentially accelerating panel aging, shortening their service life. Therefore, regular cleaning of photovoltaic panels is crucial to maintaining optimal power generation performance and extending their service life.

[0003] However, in existing photovoltaic power station cleaning technologies, cleaning plans are often based on experience or fixed schedules. This fixed cleaning plan may not only lead to excessively high or low cleaning frequencies, affecting power generation and operation and maintenance efficiency, but may also cause waste of water resources and environmental pollution. Summary of the Invention

[0004] This application aims to propose a photovoltaic module cleaning warning method, device, equipment and storage medium, which can improve the timeliness and accuracy of cleaning warnings, while ensuring the power generation, operation and maintenance efficiency and cost of photovoltaic power stations.

[0005] According to the first embodiment of the present application, a photovoltaic module cleaning early warning method includes:

[0006] Obtaining a current cleaning cycle according to a cleaning cycle model, wherein the cleaning cycle model is established based on cleaning costs and power generation losses of a target photovoltaic module;

[0007] Obtaining the last cleaning date, and determining the current cleaning date based on the current cleaning cycle and the last cleaning date;

[0008] Obtaining current panel image data of a target photovoltaic module and inputting the data into a pre-built cleanliness model to output the current cleanliness of the target photovoltaic module, wherein the cleanliness is used to indicate the cleanliness state of the surface of the target photovoltaic module, and the cleanliness model is used to obtain the corresponding cleanliness based on the input panel image data;

[0009] Obtaining the current actual power generation efficiency of the target photovoltaic assembly and inputting the result into a pre-established specific power generation efficiency model to output the current specific power generation efficiency of the target photovoltaic assembly, wherein the specific power generation efficiency is used to characterize the actual power generation capacity of the target photovoltaic assembly, and the specific power generation efficiency model is used to obtain the corresponding specific power generation efficiency based on the input actual power generation efficiency;

[0010] Obtaining a cleanliness threshold and a specific power generation efficiency threshold according to the current cleaning date, the cleanliness model, and the specific power generation efficiency model;

[0011] When the current cleanliness of the target photovoltaic component is lower than the cleanliness threshold, and / or the current specific power generation efficiency is lower than the specific power generation efficiency threshold, a cleaning warning is triggered.

[0012] According to some embodiments of the present application, the cleaning cycle model is constrained by the following expression:

[0013]

[0014] Among them, M1 is the cleaning cost of a single photovoltaic panel, N is the total number of photovoltaic panels of the target photovoltaic assembly, M0 is the cost of connecting the power generation of the power station to the grid, is the theoretical power generation on day i, H i is the effective power generation time of the i-th day, α is the daily power generation loss rate determined according to the cleanliness, and T0 is the current cleaning cycle.

[0015] According to some embodiments of the present application, obtaining the cleanliness threshold according to the current cleaning date, the cleanliness model, and the specific power generation efficiency model includes:

[0016] Acquire multiple historical panel image data of the target photovoltaic assembly, and input the cleanliness model to obtain multiple historical cleanliness values ​​corresponding to the multiple historical panel image data;

[0017] Establishing a cleanliness curve model showing changes in the cleanliness of the target photovoltaic module over time based on the current cleanliness of the target photovoltaic module and the multiple historical cleanliness levels corresponding to the multiple historical panel image data, wherein the cleanliness curve model is used to predict the future cleanliness of the target photovoltaic module;

[0018] According to the cleanliness curve model, the cleanliness corresponding to the current cleaning date is determined as a cleanliness threshold.

[0019] According to some embodiments of the present application, the cleanliness model is obtained by the following steps:

[0020] Acquiring multiple simulated panel image data of the target photovoltaic component under multiple simulated dust accumulation levels to obtain a simulated training data set;

[0021] constructing an intermediate cleanliness model based on the simulated training data set based on an image recognition algorithm;

[0022] Acquire the plurality of historical panel image data of the target photovoltaic assembly to obtain a real training data set;

[0023] The intermediate cleanliness model is trained according to the real training data set to obtain the cleanliness model.

[0024] According to some embodiments of the present application, obtaining the specific power generation efficiency threshold according to the current cleaning date, the cleanliness model, and the specific power generation efficiency model includes:

[0025] Acquire multiple historical actual power generation efficiencies of the target photovoltaic assembly and input them into the specific power generation efficiency model to obtain multiple historical specific power generation efficiencies corresponding to the multiple historical actual power generation efficiencies;

[0026] Establishing a specific power generation efficiency curve model showing the change of the specific power generation efficiency of the target photovoltaic assembly over time based on the current actual power generation efficiency of the target photovoltaic assembly and the multiple historical specific power generation efficiencies corresponding to the multiple historical actual power generation efficiencies, wherein the specific power generation efficiency curve model is used to predict the future specific power generation efficiency of the target photovoltaic assembly;

[0027] According to the specific power generation efficiency curve model, the specific power generation efficiency corresponding to the current cleaning date is determined as a specific power generation efficiency threshold.

[0028] According to some embodiments of the present application, the specific power generation efficiency model is limited by the following expression:

[0029]

[0030] in, is the specific power generation efficiency, is the actual power generation efficiency, For ideal power generation efficiency.

[0031] According to some embodiments of the present application, triggering a cleaning warning when the cleanliness of the target photovoltaic assembly is lower than the cleanliness threshold and / or the current specific power generation efficiency is lower than the specific power generation efficiency threshold includes:

[0032] When the cleanliness of the target photovoltaic module is lower than a cleanliness threshold, and / or the current specific power generation efficiency is lower than a specific power generation efficiency threshold, obtaining current rainfall data;

[0033] When the current rainfall data is less than the rainfall threshold, triggering the cleaning warning;

[0034] When the current rainfall data is greater than the rainfall threshold, the cleaning cycle is reacquired.

[0035] According to the photovoltaic module cleaning early warning device of the second embodiment of the present application, the device includes:

[0036] a current cleaning cycle acquisition module, configured to acquire the current cleaning cycle according to a cleaning cycle model, wherein the cleaning cycle model is established based on the cleaning cost and power generation loss of the target photovoltaic module;

[0037] a current cleaning date determination module, configured to obtain a last cleaning date and determine a current cleaning date based on the current cleaning cycle and the last cleaning date, wherein the last cleaning date is the date on which the target photovoltaic module was most recently cleaned;

[0038] a cleanliness acquisition module, configured to acquire current panel image data of a target photovoltaic module and input the data into a pre-built cleanliness model to output the current cleanliness of the target photovoltaic module, wherein the cleanliness is used to indicate the cleanliness state of the surface of the target photovoltaic module, and the cleanliness model is used to acquire the corresponding cleanliness based on the input panel image data;

[0039] a specific power generation efficiency acquisition module, configured to acquire the current actual power generation efficiency of the target photovoltaic assembly and input the acquired specific power generation efficiency into a pre-established specific power generation efficiency model to output the current specific power generation efficiency of the target photovoltaic assembly, wherein the specific power generation efficiency is used to characterize the actual power generation capacity of the target photovoltaic assembly, and the specific power generation efficiency model is used to acquire the corresponding specific power generation efficiency based on the input actual power generation efficiency;

[0040] a threshold determination module, configured to obtain a cleanliness threshold and a specific power generation efficiency threshold according to the current cleaning date, the cleanliness model, and the specific power generation efficiency model;

[0041] The cleaning warning module is configured to trigger a cleaning warning when the current cleanliness of the target photovoltaic component is lower than the cleanliness threshold and / or the current specific power generation efficiency is lower than the specific power generation efficiency threshold.

[0042] According to an electronic device of an embodiment of the third aspect of the present application, the device includes a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the steps of the photovoltaic component cleaning early warning method as described in any one of the embodiments of the first aspect are implemented.

[0043] According to the computer-readable storage medium of the fourth embodiment of the present application, computer-executable instructions are stored, and the computer-executable instructions are used to execute the photovoltaic component cleaning early warning method as described in the first embodiment above.

[0044] In an embodiment of the present application, a cleaning cycle model is established based on the cleaning cost and power generation loss of the target photovoltaic module to obtain the current cleaning cycle that minimizes the sum of the cleaning cost and power generation loss of the target photovoltaic module. The cleanliness and power generation efficiency of the photovoltaic module are comprehensively considered as two key factors. Based on the current cleaning cycle, a warning threshold value related to the cleanliness and power generation efficiency is determined. Subsequently, the two key factors of cleanliness and power generation efficiency are monitored in real time. When the cleanliness and power generation efficiency fall below the relevant warning threshold value, a cleaning warning is issued. This warning method can adjust the warning time according to the actual pollution situation, which not only improves the timeliness and accuracy of the cleaning warning, but also effectively avoids the loss of power generation efficiency due to excessive or insufficient cleaning. It can bring significant benefits to the operation and maintenance of photovoltaic power stations, ensure power generation, operation and maintenance efficiency and cost, and save water resources and reduce environmental pollution.

[0045] Other features and advantages of the present application will be set forth in the following description, and in part will be apparent from the description, or may be learned by practicing the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0047] Figure 1 This is a flow chart of an embodiment of the photovoltaic module cleaning early warning method of the present application;

[0048] Figure 2 is a schematic diagram of cleaning costs and power generation losses of a target photovoltaic module according to an embodiment of the present application;

[0049] Figure 3 1 is a schematic diagram of the cleanliness model construction process of an embodiment of the present application;

[0050] Figure 4 Schematic diagram of the image recognition algorithm structure of an embodiment of the present application;

[0051] Figure 5 is a schematic diagram of another image recognition algorithm structure of an embodiment of the present application;

[0052] Figure 6 is a schematic diagram of a specific power generation efficiency curve of an embodiment of the present application;

[0053] Figure 7 This is a schematic structural diagram of an embodiment of a photovoltaic module cleaning warning device of the present application;

[0054] Figure 8 It is a hardware structure diagram of an embodiment of the electronic device of the present application. DETAILED DESCRIPTION

[0055] The following describes in detail embodiments of the present application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application and are not to be construed as limiting the present application.

[0056] In the description of this application, if there is a description of first, second, etc., it is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0057] In the description of this application, it should be understood that descriptions involving orientation, such as the orientation or positional relationship indicated by up, down, etc., are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on this application.

[0058] In the description of this application, it should be noted that, unless otherwise clearly defined, terms such as setting, installing, and connecting should be understood in a broad sense, and technical personnel in the relevant technical field can reasonably determine the specific meaning of the above terms in this application based on the specific content of the technical solution.

[0059] The technical solution of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the embodiments described below are only part of the embodiments of the present application, not all of the embodiments.

[0060] Figure 1 A schematic diagram of a process flow of a photovoltaic module cleaning early warning method according to an embodiment of the present application; Figure 2 This is a schematic structural diagram of an embodiment of a photovoltaic module cleaning warning device of the present application; Figure 3 It is a hardware structure diagram of an embodiment of the electronic device of the present application.

[0061] See below Figure 1 , further elaborating on the embodiments of this application.

[0062] The present application provides a photovoltaic module cleaning early warning method, which includes the following steps:

[0063] Step 101: Obtain a current cleaning cycle according to a cleaning cycle model, where the cleaning cycle model is established based on the cleaning cost and power generation loss of a target photovoltaic module;

[0064] Step 102: Obtain the last cleaning date and determine the current cleaning date based on the current cleaning cycle and the last cleaning date.

[0065] Step 103: obtaining current panel image data of the target photovoltaic module and inputting the data into a pre-built cleanliness model to output the current cleanliness of the target photovoltaic module, wherein the cleanliness is used to indicate the cleanliness state of the surface of the target photovoltaic module, and the cleanliness model is used to obtain the corresponding cleanliness based on the input panel image data;

[0066] Step 104: Obtain the current actual power generation efficiency of the target photovoltaic module and input it into a pre-established specific power generation efficiency model to output the current specific power generation efficiency of the target photovoltaic module. The specific power generation efficiency is used to characterize the actual power generation capacity of the target photovoltaic module, and the specific power generation efficiency model is used to obtain the corresponding specific power generation efficiency based on the input actual power generation efficiency.

[0067] Step 105: Obtain a cleanliness threshold and a specific power generation efficiency threshold based on the current cleaning date, the cleanliness model, and the specific power generation efficiency model;

[0068] Step 106 : When the current cleanliness of the target photovoltaic module is lower than a cleanliness threshold, and / or the current specific power generation efficiency is lower than a specific power generation efficiency threshold, trigger a cleaning warning.

[0069] In an embodiment of the present application, a cleaning cycle model is established based on the cleaning cost and power generation loss of the target photovoltaic module to obtain the current cleaning cycle that minimizes the sum of the cleaning cost and power generation loss of the target photovoltaic module. The cleanliness and power generation efficiency of the photovoltaic module are comprehensively considered. Based on the current cleaning cycle, a warning threshold value related to the cleanliness and power generation efficiency is determined. Subsequently, the two key factors of cleanliness and power generation efficiency are monitored in real time. When the cleanliness and power generation efficiency fall below the relevant warning threshold value, a cleaning warning is issued. This warning method can adjust the warning time according to the actual pollution situation, which not only improves the timeliness and accuracy of cleaning, but also effectively avoids the loss of power generation efficiency due to excessive or insufficient cleaning. It can bring significant benefits to the operation and maintenance of photovoltaic power stations, ensure power generation and operation and maintenance efficiency, and save water resources and reduce environmental pollution.

[0070] The above-mentioned last cleaning date is the date on which the target photovoltaic module was actually cleaned in the past, and the above-mentioned current cleaning date is the best ideal date on which the target photovoltaic module is expected to be cleaned in the future, and is also the best cleaning date that minimizes the sum of the cleaning cost and power generation loss of the target photovoltaic module.

[0071] It should be noted that the cleaning date is not necessarily the date of the most recent actual cleaning in the future. This cleaning date is only an estimated value, and this estimated value is used to determine the cleanliness threshold and the power generation efficiency threshold. Finally, the cleanliness threshold and the power generation efficiency threshold are used to determine whether to trigger the cleaning warning.

[0072] After the above cleaning warning is triggered, a red warning prompt will be issued in a prominent position on the front-end page to remind staff to arrange cleaning tasks as soon as possible to avoid continuous decline in power generation performance due to pollution accumulation.

[0073] The above-mentioned panel image data can be obtained periodically. The camera or other image acquisition equipment deployed at the photovoltaic power station site can automatically perform image capture operations at preset time points or time intervals to obtain real-time image information of the surface of the photovoltaic module and store it on the server.

[0074] Specifically, scheduled acquisition of panel image data involves automatically capturing images at preset time points or intervals using cameras or other image acquisition devices deployed at the PV power plant site to obtain real-time image information of the PV module surface. This process requires no human intervention; the system's backend control module triggers image acquisition tasks based on a predefined acquisition strategy. Typically, three acquisition times are set for the morning and afternoon. For example, in the morning, images can be collected at 8:00, 10:00, and 12:00, and in the afternoon at 14:00, 16:00, and 18:00. This configuration helps capture the module surface conditions under different solar altitudes and lighting conditions, improving the image recognition model's adaptability to environmental changes. Image data storage involves storing the captured image data on a server in JPEG format. The storage file format uses a multi-layered naming structure. The server stores files for each camera, and each camera stores files for each date.

[0075] The above cleanliness is limited by the following expression:

[0076]

[0077] Among them, Cleanliness is cleanliness, P actual is the actual generated power, P clean The cleanliness is the power generation capacity in a clean state. The cleanliness can reflect the cleanliness of the component surface. It is the ratio of the actual power generation capacity to the power generation capacity in a clean state. The value range is 0 to 1. The closer the value is to 1, the cleaner it is, and the closer to 0, the dirtier it is.

[0078] In some embodiments, the cleaning cycle model is constrained by the following expression:

[0079]

[0080] Among them, M1 is the cleaning cost of a single photovoltaic panel, N is the total number of photovoltaic panels in the target photovoltaic module, M0 is the cost of connecting the power generation of the power station to the grid, is the theoretical power generation on day i, H i is the effective power generation time on the i-th day, α is the daily power generation loss rate determined by the cleanliness, and T0 is the current cleaning cycle.

[0081] In this embodiment, when the power generation loss is equal to the cleaning cost, the sum of the cleaning cost and the power generation loss of the target photovoltaic module can be minimized, and the current cleaning cycle determined is the theoretically optimal cleaning cycle.

[0082] like Figure 2 The figure shows the calculation process for the cleaning cycle to minimize cost. The objective function for minimizing the cost cycle is to minimize the sum of the power generation company's operating losses and cleaning costs. The optimal cleaning cycle is used as the current cleaning cycle, which is also the minimum cost cycle. In the figure, the cleaning cost C is parallel to the horizontal axis and is specifically constrained by the following expression:

[0083] C = M1 × N;

[0084] Among them, M1 is the cleaning cost of each photovoltaic panel in the photovoltaic station, and N is the total number of photovoltaic panels that need to be cleaned.

[0085] The power generation loss, L, is limited by the following expression:

[0086]

[0087] Among them, M0 is the cost of connecting the power generation of the power station to the grid, which is determined by the local electricity price policy and actual situation. is the theoretical power generation on day i, which is calculated based on the model or experimentally measured under standard irradiation conditions. i is the effective power generation time on day i, usually set to an estimated value, such as 10 hours on a sunny day, θ is the number of days, and α is the daily power loss rate determined by the cleanliness. The daily power loss rate α and the cleanliness level are constrained by the following expression:

[0088] Cleanliness = 1-α;

[0089] Specifically, the value of α can be the daily power generation loss rate value obtained based on the cleanliness data corresponding to the sunny day sample data, which is obtained by fitting a large number of sunny day sample experimental data; the value of α can also be the daily power generation loss rate value obtained by the cleanliness data corresponding to a certain stage, specifically, it can be determined as the daily power generation loss rate value obtained by the cleanliness data corresponding to the previous cleaning cycle or the current cleaning cycle. The value of α will be a dynamic value that changes with time, and thus the cleaning cycle T0 will also be a dynamic value.

[0090] In some cases, α can also be determined based on the solar radiation. The daily power generation loss rate α and the solar radiation G per unit time have the following approximate relationship:

[0091]

[0092] Where ΔE is the power generation loss caused by dust accumulation, in kWh, and G is the effective total solar radiation on that day, in kWh / m 2 , η is the module conversion efficiency, and A is the module area. Therefore, when the actual sunshine duration increases with seasonal solar radiation, the cleaning cycle will also decrease; similarly, when the actual sunshine duration decreases, the cleaning cycle will also decrease.

[0093] When the power generation loss is equal to the cleaning cost, that is, when L and C are equal, the value of the number of days θ can be obtained and determined as the current cleaning cycle T0, which is the optimal cleaning time point.

[0094] The current cleaning cycle T0 is the time period from the last cleaning to the dust accumulation on the component surface until the next cleaning begins. The cleaning process needs to continue for a certain period of time, that is, the cleaning process duration T k The overall cleaning cycle T of the photovoltaic module is the duration of the dust accumulation process, that is, T0, and the duration of the cleaning process, T k The sum is:

[0095] T=T0+T k ;

[0096] And T and T0 satisfy the following expressions:

[0097]

[0098] Therefore, the overall cleaning cycle T of the photovoltaic module is limited by the following expression:

[0099]

[0100] Among them, M1 is the cleaning cost of a single photovoltaic panel at a photovoltaic station, which is derived from historical cleaning cost data; N is the total number of photovoltaic panels that need to be cleaned, which is provided by the specific station; M0 is the cost of connecting the power generation to the grid, which is determined by the local electricity price policy and actual conditions; W Li is the theoretical power generation on day i, which is calculated based on the model or measured experimentally under standard irradiation conditions; H i is the effective power generation time on the i-th day, which is usually set to an estimated value, such as 10 hours on a sunny day; α is the daily power generation loss rate; n′ is the actual operating days of all components in a certain stage, which is calculated from operating records or experimental observation data; T is the optimal overall cleaning cycle, which is the target parameter to be solved, aiming to achieve the optimal balance between cleaning cost and power generation efficiency.

[0101] In some cases, the cleaning process duration T can also be preset based on historical operation data or cleaning scheduling plan. k , referring to the actual situation of a power station in Northwest China, it is generally about 30 days. Therefore, to obtain the best overall cleaning cycle, we only need to get T0 plus T k You can get it.

[0102] In some embodiments, the last cleaning date is obtained, and the current cleaning date is determined based on the current cleaning cycle and the last cleaning date. This can be done by adding the cleaning process duration T to the last cleaning date. k , plus the current cleaning cycle T0, the cleaning date can be determined.

[0103] In some embodiments, obtaining a cleanliness threshold according to the current cleaning date, the cleanliness model, and the specific power generation efficiency model includes:

[0104] Acquire multiple historical panel image data of a target photovoltaic module, input the data into a cleanliness model, and obtain multiple historical cleanliness values ​​corresponding to the multiple historical panel image data;

[0105] Based on the current cleanliness of the target photovoltaic module and multiple historical cleanliness values ​​corresponding to multiple historical panel image data, a cleanliness curve model is established to show how the cleanliness of the target photovoltaic module changes over time. The cleanliness curve model is used to predict the future cleanliness of the target photovoltaic module.

[0106] According to the cleanliness curve model, the cleanliness corresponding to this cleaning date is determined as the cleanliness threshold.

[0107] In this implementation, based on the continuous processing of long-term data, a cleanliness curve can be plotted, showing the changes in PV panel cleanliness over time, based on multiple historical cleanliness levels corresponding to multiple historical panel image data. This cleanliness curve model can be used to dynamically monitor the cleanliness status of the panel surface and, to a certain extent, fit future cleanliness data, providing data support for determining cleaning timing and optimizing cleaning strategies. Therefore, based on the current cleaning date, the cleanliness level corresponding to that cleaning date in the cleanliness curve model can be determined as the cleanliness threshold, thereby obtaining a more appropriate judgment threshold.

[0108] The above-mentioned historical panel image data is the panel image data collected before obtaining the current panel image data. It should be noted that data within a historical time period can be used, and all historical data can also be used.

[0109] In some embodiments, the cleanliness threshold is determined based on a cleanliness curve model. To ensure the scientific nature of the threshold, a weighted average of multiple cleanliness values ​​before and after the current cleaning date can be taken to determine the final cleanliness threshold.

[0110] In some embodiments, the cleanliness model is obtained by the following steps:

[0111] Acquire multiple simulated panel image data of a target photovoltaic module under multiple simulated dust accumulation levels to obtain a simulated training data set;

[0112] Based on the image recognition algorithm, an intermediate cleanliness model is constructed according to the simulated training data set;

[0113] Acquire multiple historical panel image data of the target photovoltaic module to obtain a real training data set;

[0114] According to the real training data set, the intermediate cleanliness model is trained to obtain the cleanliness model.

[0115] In this embodiment, the cleanliness model is constructed based on an image recognition algorithm, mainly by performing image classification and dust accumulation degree assessment on photovoltaic module image data collected by a camera at regular intervals.

[0116] The above acquisition of multiple simulated panel image data of the target PV module at various simulated dust accumulation levels generates a simulated training dataset, which can be used for dust accumulation simulation experiments. The dust accumulation simulation experiment is based on the on-site scenario simulation of the PV station. It simulates the full range of contamination levels of the PV panel surface, from clean to heavily dusted. It uses manual dust spraying and introduces a small amount of water droplets to replicate the dew adhesion effect, generating dust layers with different spatial distribution densities. Finally, multi-angle, high-resolution images are captured. The image dataset will cover different levels of dust coverage, providing multi-dimensional benchmark data support for the subsequent deep learning-based dust detection algorithm training and cleaning demand prediction model development.

[0117] In some embodiments, the simulation data set is first classified according to the dust accumulation degree of the photovoltaic panel in the dust accumulation simulation experiment, and then the simulation data set is trained using an image recognition algorithm. The model is further verified using the acquired field images. The model is trained according to the image recognition algorithm, and then the field image data is substituted to extract the feature information related to dust accumulation. Then, according to the classification result, the corresponding grayscale value is mapped out. Further, the grayscale is converted into a cleanliness value through a formula. It should be noted that the cleanliness is equal to 1 minus the grayscale, that is, the grayscale is numerically equal to the daily power generation loss rate.

[0118] In some embodiments, as Figure 3As shown in the figure, a gray accumulation simulation experiment is first carried out to divide the dust accumulation on the surface of the photovoltaic module panel into 15 levels, such as 63%, 68%, 73% to 100%, and the corresponding power generation value is measured at different dust accumulation levels to establish a preliminary correspondence between power loss and dust accumulation degree. Subsequently, the dust accumulation image dataset is constructed using the dust accumulation level corresponding images collected during the experiment, and it is input as a training sample into the image recognition algorithm Resnet50 to train the model to extract the dust accumulation features in the image and output the simulated dust accumulation degree value, which is recorded as the simulated gray accumulation degree.

[0119] Building on this foundation, a large amount of previously collected on-site PV panel image data stored on the server was further utilized and fed into the same image recognition network for training and optimization. Because these images are derived from actual PV panels in operation and have labels that match measured power data, they can be used to refine and improve model accuracy, resulting in more reliable dust accumulation predictions, recorded as the true dust accumulation level.

[0120] The final output of the model is the cleanliness value of the photovoltaic module, which is defined as:

[0121]

[0122] Among them, Cleanliness is cleanliness, P actual is the actual generated power, P clean Cleanliness represents the power generated in a clean state. Cleanliness reflects the cleanliness of the panel surface and is the ratio of actual power generation capacity to that in a clean state. Its value ranges from 0 to 1, with values ​​closer to 1 indicating cleaner conditions and closer to 0 indicating dirtier surfaces. With continuous data input and iterative model optimization, the system generates a dynamic cleanliness curve, which monitors the contamination status of PV panels in real time and provides data support for subsequent cleaning scheduling and decision-making.

[0123] like Figure 4 and Figure 5 The image recognition algorithm of the present invention adopts the Resnet deep convolutional neural network architecture, and its core idea revolves around the construction of residual blocks, such as Figure 4 As shown in the figure, the residual block embeds the input layer, convolutional layer, ReLU activation function, and output layer, a total of four groups of large modules (blocks), each group of 3, 4, 6, and 3 small modules (blocks), respectively. Each small module (block) contains three convolutions. Together with the initial single convolutional layer and the final fully connected layer, (3+4+6+3)*3+1+1=50, a total of 50 layers. The residual connection mechanism is used to directly transfer features. The four residual blocks are stacked hierarchically to form the overall architecture of ResNet 50.

[0124] like Figure 5 As shown in the figure, within each large module (block), the first small module (block) is located on the left branch, where the input (IN) is not equal to the output (OUTPUT). It is named the Convolution Block. The remaining small modules are located on the right branch, where the input is equal to the output. They are named the Identity Block. The Convolution Block uses convolutional layers to adjust the input size to match the output of the main path, primarily used for feature extraction and resizing. The Identity Block maintains the consistency of input and output dimensions through identity mapping, primarily used to increase the depth of the network. The two modules are used alternately to build more complex deep learning models.

[0125] The algorithm's workflow is as follows: The classified image data is first fed into the input layer. Then, in the residual block, the image data undergoes multiple layers of convolution operations and nonlinear transformations using the ReLU activation function, gradually extracting high-level features from the image. After processing through these residual blocks, the final feature representation is passed to the output layer, which is used to predict the image's dust accumulation levels. After training, the ResNet model can accept any input image and output its corresponding dust accumulation value, which can then be used to construct the subsequent cleanliness model.

[0126] In some embodiments, obtaining the specific power generation efficiency threshold according to the current cleaning date, the cleanliness model, and the specific power generation efficiency model includes:

[0127] Acquire multiple historical actual power generation efficiencies of the target photovoltaic module and input them into a specific power generation efficiency model to obtain multiple historical specific power generation efficiencies corresponding to the multiple historical actual power generation efficiencies;

[0128] Based on the current actual power generation efficiency of the target photovoltaic module and multiple historical specific power generation efficiencies corresponding to the multiple historical actual power generation efficiencies, a specific power generation efficiency curve model is established to show the change of the specific power generation efficiency of the target photovoltaic module over time. The specific power generation efficiency curve model is used to predict the future specific power generation efficiency of the target photovoltaic module.

[0129] According to the specific power generation efficiency curve model, the specific power generation efficiency corresponding to the current cleaning date is determined as the specific power generation efficiency threshold.

[0130] In this embodiment, based on the continuous processing of long-term data, a specific power generation efficiency curve can be plotted, showing the change in the specific power generation efficiency of a PV module over time, based on the multiple historical specific power generation efficiencies corresponding to multiple historical actual power generation efficiencies. This specific power generation efficiency curve model can be used to dynamically monitor the power generation efficiency status of the module and, to a certain extent, fit future specific power generation efficiency data, providing data support for determining cleaning timing and optimizing cleaning strategies. Therefore, based on the determined current cleaning date, the specific power generation efficiency corresponding to the current cleaning date in the specific power generation efficiency curve model is determined as the specific power generation efficiency threshold, thereby obtaining a relatively appropriate judgment threshold.

[0131] The above historical actual power generation efficiency is the actual power generation efficiency obtained before obtaining the current actual power generation efficiency. It should be noted that data within a historical time period can be used, and all historical data can also be used.

[0132] In some embodiments, the relative power generation efficiency threshold is determined based on the relative power generation efficiency curve model. To ensure the scientific nature of the threshold, a weighted average of multiple relative power generation efficiency values ​​for a period of time before and after the current cleaning date can be taken to determine the final relative power generation efficiency threshold.

[0133] In some embodiments, the specific power generation efficiency model is constrained by the following expression:

[0134]

[0135] in, is the specific power generation efficiency, is the actual power generation efficiency, For ideal power generation efficiency.

[0136] In this embodiment, is the specific power generation efficiency of photovoltaic modules, The ideal power generation efficiency is set to the power generation efficiency of the PV module after cleaning, which is numerically equal to the ratio of the power generation after cleaning to the power generation under ideal conditions; The actual power generation efficiency is the power generation efficiency of the photovoltaic module in the current state, which is numerically equal to the ratio of the power generation in the current state to the power generation in the ideal state. Subtract actual power generation efficiency Divide by the ideal power generation efficiency This reflects the impact of dust accumulation on the module's power generation efficiency. Substituting site-related data into this formula yields a specific power generation efficiency curve model.

[0137] The above-mentioned specific power generation efficiency reflects the actual power generation capacity of photovoltaic modules.

[0138] In some embodiments, based on the current actual power generation efficiency of the target photovoltaic component and multiple historical relative power generation efficiencies corresponding to multiple historical actual power generation efficiencies, a relative power generation efficiency curve model of the target photovoltaic component changing with time is established, and the relative power generation efficiency curve model is used to predict the future relative power generation efficiency of the target photovoltaic component.

[0139] like Figure 6 The figure below shows a schematic diagram of the specific power generation efficiency curve. This fitted curve is useful for predicting future specific power generation efficiency. The steps for establishing the fitted curve are: subtract the actual power generation efficiency from the ideal power generation efficiency corresponding to multiple sets of data, then divide by the ideal power generation efficiency to obtain the degree of impact of dust accumulation on module power generation. This data is then substituted into the specific power generation efficiency model formula to form a scatter plot. The slope k of the line in the regression model and the intercept b on the vertical axis are obtained by fitting the scatter plot. This provides the linear relationship between the actual module power generation efficiency and the duration of dust accumulation under the same environmental conditions.

[0140] In some cases, the parameters in the regression model are derived from data from the most recent cleaning cycle, resulting in a variable slope. Understandably, as irradiance increases from June to August, the impact of other factors increases, and the difference between actual and theoretical power generation decreases, leading to a smaller slope. This means that the impact of dust accumulation on PV module efficiency approaches its limit. Furthermore, the amount of dust accumulated on the PV panel surface is related to the duration of previous accumulation. Dust accumulation is fastest in the first few days, resulting in a greater impact on power generation, and then gradually decreases toward a flattening trend.

[0141] In some embodiments, when the cleanliness of a target photovoltaic module is lower than a cleanliness threshold, and / or the current specific power generation efficiency is lower than a specific power generation efficiency threshold, triggering a cleaning warning includes:

[0142] When the cleanliness of the target photovoltaic module is lower than a cleanliness threshold, and / or the current specific power generation efficiency is lower than a specific power generation efficiency threshold, obtaining current rainfall data;

[0143] When the current rainfall data is less than the rainfall threshold, a cleaning warning is triggered;

[0144] When the current rainfall data is greater than the rainfall threshold, the cleaning cycle is re-acquired.

[0145] In this embodiment, the impact of rainfall on the cleaning cycle is also taken into account. When the current cleanliness and / or power generation efficiency drops below the threshold, if the current rainfall data is greater than the rainfall threshold, it can be considered that the rainfall will complete a cleaning cycle and enter the next overall cleaning cycle stage. If the current rainfall data is less than the rainfall threshold, a cleaning warning still needs to be triggered.

[0146] The above-mentioned current rainfall data may be rainfall data within the current 24 hours, or rainfall data within the next seven days. The above-mentioned rainfall threshold may be preset according to actual conditions.

[0147] In some embodiments, when a cleaning warning is about to be triggered in the near future, if there is currently rainfall, the system will automatically update the cleaning warning cycle according to the actual situation, and the cleaning warning countdown will be automatically reset. The entire process does not require manual intervention; if there is currently no rainfall, but there is a rainfall forecast within the next seven days, the system will dynamically evaluate whether to perform the cleaning operation based on the forecast accuracy, rainfall intensity and equipment operating status to achieve optimal resource allocation and operation and maintenance cost control. Specifically, if the current cleanliness and power generation efficiency curves are both lower than the warning values, but the rainfall in the next seven days is greater than or equal to the set rainfall value, the system will issue a warning to remind the site to pay attention, but there is no need to perform cleaning operations. If the rainfall in the next seven days is less than the set rainfall value, the system will issue a warning to remind the site to perform cleaning operations. If there is no rainfall currently and in the next seven days, the system will monitor the component cleanliness and power generation efficiency indicators in real time. If both parameters are lower than the preset warning thresholds, a cleaning warning will be triggered.

[0148] The photovoltaic module cleaning warning method provided in the embodiment of the present application can be executed by the photovoltaic module cleaning warning device 200. In the embodiment of the present application, the photovoltaic module cleaning warning device 200 is used as an example to illustrate the photovoltaic module cleaning warning method provided in the embodiment of the present application.

[0149] See Figure 7 , is a schematic diagram of the structure of a photovoltaic module cleaning warning device 200 provided in an embodiment of the present application. Figure 7 As shown, the photovoltaic module cleaning early warning device 200 includes:

[0150] According to the photovoltaic module cleaning early warning device of the second embodiment of the present application, the device includes:

[0151] The current cleaning cycle acquisition module 201 is used to acquire the current cleaning cycle according to a cleaning cycle model, where the cleaning cycle model is established based on the cleaning cost and power generation loss of the target photovoltaic module;

[0152] The current cleaning date determination module 202 is used to obtain the last cleaning date and determine the current cleaning date based on the current cleaning cycle and the last cleaning date, where the last cleaning date is the date when the target photovoltaic module was most recently cleaned;

[0153] Cleanliness acquisition module 203 is used to obtain the current panel image data of the target photovoltaic module and input it into a pre-built cleanliness model to output the current cleanliness of the target photovoltaic module, wherein the cleanliness is used to indicate the cleanliness state of the surface of the target photovoltaic module, and the cleanliness model is used to obtain the corresponding cleanliness based on the input panel image data;

[0154] The specific power generation efficiency acquisition module 204 is used to obtain the current actual power generation efficiency of the target photovoltaic module and input it into a pre-established specific power generation efficiency model to output the current specific power generation efficiency of the target photovoltaic module. The specific power generation efficiency is used to represent the actual power generation capacity of the target photovoltaic module, and the specific power generation efficiency model is used to obtain the corresponding specific power generation efficiency based on the input actual power generation efficiency.

[0155] A threshold determination module 205 is configured to obtain a cleanliness threshold and a specific power generation efficiency threshold based on the current cleaning date, the cleanliness model, and the specific power generation efficiency model;

[0156] The cleaning warning module 206 is configured to trigger a cleaning warning when the current cleanliness of the target photovoltaic module is lower than a cleanliness threshold and / or the current specific power generation efficiency is lower than a specific power generation efficiency threshold.

[0157] In some embodiments, in the current cleaning cycle acquisition module 201, the cleaning cycle model is constrained by the following expression:

[0158]

[0159] Among them, M1 is the cleaning cost of a single photovoltaic panel, N is the total number of photovoltaic panels in the target photovoltaic module, M0 is the cost of connecting the power generation of the power station to the grid, is the theoretical power generation on day i, H i is the effective power generation time on the i-th day, α is the daily power generation loss rate determined by the cleanliness, and T0 is the current cleaning cycle.

[0160] In some implementations, the threshold determination module 205 may be configured to:

[0161] Acquire multiple historical panel image data of a target photovoltaic module, input the data into a cleanliness model, and obtain multiple historical cleanliness values ​​corresponding to the multiple historical panel image data;

[0162] Based on the current cleanliness of the target photovoltaic module and multiple historical cleanliness values ​​corresponding to multiple historical panel image data, a cleanliness curve model is established to show how the cleanliness of the target photovoltaic module changes over time. The cleanliness curve model is used to predict the future cleanliness of the target photovoltaic module.

[0163] According to the cleanliness curve model, the cleanliness corresponding to this cleaning date is determined as the cleanliness threshold.

[0164] In some embodiments, in the cleanliness acquisition module 203, the cleanliness model is obtained by the following steps:

[0165] Acquire multiple simulated panel image data of a target photovoltaic module under multiple simulated dust accumulation levels to obtain a simulated training data set;

[0166] Based on the image recognition algorithm, an intermediate cleanliness model is constructed according to the simulated training data set;

[0167] Acquire multiple historical panel image data of the target photovoltaic module to obtain a real training data set;

[0168] According to the real training data set, the intermediate cleanliness model is trained to obtain the cleanliness model.

[0169] In some implementations, the threshold determination module 205 may be configured to:

[0170] Acquire multiple historical actual power generation efficiencies of the target photovoltaic module and input them into a specific power generation efficiency model to obtain multiple historical specific power generation efficiencies corresponding to the multiple historical actual power generation efficiencies;

[0171] Based on the current actual power generation efficiency of the target photovoltaic module and multiple historical specific power generation efficiencies corresponding to the multiple historical actual power generation efficiencies, a specific power generation efficiency curve model is established to show the change of the specific power generation efficiency of the target photovoltaic module over time. The specific power generation efficiency curve model is used to predict the future specific power generation efficiency of the target photovoltaic module.

[0172] According to the specific power generation efficiency curve model, the specific power generation efficiency corresponding to the current cleaning date is determined as the specific power generation efficiency threshold.

[0173] In some embodiments, in the specific power generation efficiency acquisition module 204, the specific power generation efficiency model is constrained by the following expression:

[0174]

[0175] in, is the specific power generation efficiency, is the actual power generation efficiency, For ideal power generation efficiency.

[0176] In some embodiments, the cleaning warning module 206 can be used to:

[0177] When the cleanliness of the target photovoltaic module is lower than a cleanliness threshold, and / or the current specific power generation efficiency is lower than a specific power generation efficiency threshold, obtaining current rainfall data;

[0178] When the current rainfall data is less than the rainfall threshold, a cleaning warning is triggered;

[0179] When the current rainfall data is greater than the rainfall threshold, the cleaning cycle is retrieved.

[0180] Since the photovoltaic module cleaning warning device 200 adopts all the technical solutions of the photovoltaic module cleaning warning method of the above embodiment, it has at least all the beneficial effects brought by the technical solutions of the above embodiment, which will not be described in detail here.

[0181] Figure 8 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application.

[0182] The electronic device may include a processor 301 and a memory 302 storing computer program instructions.

[0183] Specifically, the processor 301 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0184] The memory 302 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 302 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 302 may include removable or non-removable (or fixed) media. Where appropriate, the memory 302 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 302 is a non-volatile solid-state memory.

[0185] In some embodiments, the memory 302 may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical / tangible memory storage devices. Thus, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0186] The processor 301 reads and executes computer program instructions stored in the memory 302 to implement any one of the photovoltaic module cleaning early warning methods in the above embodiments.

[0187] In one example, the electronic device may further include a communication interface 303 and a bus 310. Figure 8 As shown, the processor 301 , the memory 302 , and the communication interface 303 are connected via a bus 310 and communicate with each other.

[0188] The communication interface 303 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0189] Bus 310 includes hardware, software or both, and the components of online data flow metering equipment are coupled to each other. For example, but not limitation, bus can include accelerated graphics port (AGP) or other graphics bus, enhanced industry standard architecture (EISA) bus, front side bus (FSB), hypertransport (HT) interconnection, industry standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 310 can include one or more buses. Although the present application embodiment describes and shows specific bus, the application considers any suitable bus or interconnection.

[0190] The electronic device can execute the photovoltaic module cleaning warning method in the embodiment of the present application, thereby realizing the combination Figure 1 and Figure 7 The invention describes a photovoltaic module cleaning early warning method and device.

[0191] In addition, in conjunction with the photovoltaic module cleaning warning method in the above embodiments, embodiments of the present application may provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any of the photovoltaic module cleaning warning methods in the above embodiments is implemented.

[0192] It should be understood that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present application is not limited to the specific steps described and illustrated. Those skilled in the art can make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present application.

[0193] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, they can be electronic circuits, application specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROMs, flash memories, erasable ROMs (EROMs), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0194] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps. In other words, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0195] Aspects of the present disclosure have been described above with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit. It is also understood that each box in the block diagram and / or flowchart and the combination of the boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0196] The above description is only a specific embodiment of the present application. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the scope of protection of the present application.

Claims

1. A photovoltaic module cleaning early warning method, characterized in that: include: Obtaining a current cleaning cycle according to a cleaning cycle model, wherein the cleaning cycle model is established based on cleaning costs and power generation losses of a target photovoltaic module; Obtaining the last cleaning date, and determining the current cleaning date based on the current cleaning cycle and the last cleaning date; Obtaining current panel image data of a target photovoltaic module and inputting the data into a pre-built cleanliness model to output the current cleanliness of the target photovoltaic module, wherein the cleanliness is used to indicate the cleanliness state of the surface of the target photovoltaic module, and the cleanliness model is used to obtain the corresponding cleanliness based on the input panel image data; Obtaining the current actual power generation efficiency of the target photovoltaic assembly and inputting the result into a pre-established specific power generation efficiency model to output the current specific power generation efficiency of the target photovoltaic assembly, wherein the specific power generation efficiency is used to characterize the actual power generation capacity of the target photovoltaic assembly, and the specific power generation efficiency model is used to obtain the corresponding specific power generation efficiency based on the input actual power generation efficiency; Obtaining a cleanliness threshold and a specific power generation efficiency threshold according to the current cleaning date, the cleanliness model, and the specific power generation efficiency model; When the current cleanliness of the target photovoltaic component is lower than the cleanliness threshold, and / or the current specific power generation efficiency is lower than the specific power generation efficiency threshold, a cleaning warning is triggered.

2. The photovoltaic module cleaning early warning method according to claim 1, characterized in that: The cleaning cycle model is constrained by the following expression: Among them, M1 is the cleaning cost of a single photovoltaic panel, N is the total number of photovoltaic panels of the target photovoltaic assembly, M0 is the cost of connecting the power generation of the power station to the grid, is the theoretical power generation on day i, H i is the effective power generation time of the i-th day, α is the daily power generation loss rate determined according to the cleanliness, and 0 is the current cleaning cycle.

3. The photovoltaic module cleaning early warning method according to claim 1, characterized in that: Obtaining a cleanliness threshold according to the current cleaning date, the cleanliness model, and the specific power generation efficiency model includes: Acquire multiple historical panel image data of the target photovoltaic assembly, and input the cleanliness model to obtain multiple historical cleanliness values ​​corresponding to the multiple historical panel image data; Establishing a cleanliness curve model showing changes in the cleanliness of the target photovoltaic module over time based on the current cleanliness of the target photovoltaic module and the multiple historical cleanliness levels corresponding to the multiple historical panel image data, wherein the cleanliness curve model is used to predict the future cleanliness of the target photovoltaic module; According to the cleanliness curve model, the cleanliness corresponding to the current cleaning date is determined as a cleanliness threshold.

4. The photovoltaic module cleaning early warning method according to claim 1 or 3, characterized in that: The cleanliness model is obtained by the following steps: Acquiring multiple simulated panel image data of the target photovoltaic component under multiple simulated dust accumulation levels to obtain a simulated training data set; constructing an intermediate cleanliness model based on the simulated training data set based on an image recognition algorithm; Acquire the plurality of historical panel image data of the target photovoltaic assembly to obtain a real training data set; The intermediate cleanliness model is trained according to the real training data set to obtain the cleanliness model.

5. The photovoltaic module cleaning early warning method according to claim 1, characterized in that: Obtaining the specific power generation efficiency threshold according to the current cleaning date, the cleanliness model, and the specific power generation efficiency model includes: Acquire multiple historical actual power generation efficiencies of the target photovoltaic assembly and input them into the specific power generation efficiency model to obtain multiple historical specific power generation efficiencies corresponding to the multiple historical actual power generation efficiencies; Establishing a specific power generation efficiency curve model showing the change of the specific power generation efficiency of the target photovoltaic assembly over time based on the current actual power generation efficiency of the target photovoltaic assembly and the multiple historical specific power generation efficiencies corresponding to the multiple historical actual power generation efficiencies, wherein the specific power generation efficiency curve model is used to predict the future specific power generation efficiency of the target photovoltaic assembly; According to the specific power generation efficiency curve model, the specific power generation efficiency corresponding to the current cleaning date is determined as a specific power generation efficiency threshold.

6. The photovoltaic module cleaning early warning method according to claim 1 or 5, characterized in that: The specific power generation efficiency model is constrained by the following expression: in, is the specific power generation efficiency, is the actual power generation efficiency, For ideal power generation efficiency.

7. The photovoltaic module cleaning early warning method according to claim 1, characterized in that: The triggering of a cleaning warning when the cleanliness of the target photovoltaic component is lower than the cleanliness threshold and / or the current specific power generation efficiency is lower than the specific power generation efficiency threshold includes: When the cleanliness of the target photovoltaic module is lower than a cleanliness threshold, and / or the current specific power generation efficiency is lower than a specific power generation efficiency threshold, obtaining current rainfall data; When the current rainfall data is less than the rainfall threshold, triggering the cleaning warning; When the current rainfall data is greater than the rainfall threshold, the cleaning cycle is reacquired.

8. A photovoltaic module cleaning early warning device, characterized in that: include: a current cleaning cycle acquisition module, configured to acquire the current cleaning cycle according to a cleaning cycle model, wherein the cleaning cycle model is established based on the cleaning cost and power generation loss of the target photovoltaic module; a current cleaning date determination module, configured to obtain a last cleaning date and determine a current cleaning date based on the current cleaning cycle and the last cleaning date, wherein the last cleaning date is the date on which the target photovoltaic module was most recently cleaned; a cleanliness acquisition module, configured to acquire current panel image data of a target photovoltaic module and input the data into a pre-built cleanliness model to output the current cleanliness of the target photovoltaic module, wherein the cleanliness is used to indicate the cleanliness state of the surface of the target photovoltaic module, and the cleanliness model is used to acquire the corresponding cleanliness based on the input panel image data; a specific power generation efficiency acquisition module, configured to acquire the current actual power generation efficiency of the target photovoltaic assembly and input the acquired specific power generation efficiency into a pre-established specific power generation efficiency model to output the current specific power generation efficiency of the target photovoltaic assembly, wherein the specific power generation efficiency is used to characterize the actual power generation capacity of the target photovoltaic assembly, and the specific power generation efficiency model is used to acquire the corresponding specific power generation efficiency based on the input actual power generation efficiency; a threshold determination module, configured to obtain a cleanliness threshold and a specific power generation efficiency threshold according to the current cleaning date, the cleanliness model, and the specific power generation efficiency model; The cleaning warning module is configured to trigger a cleaning warning when the current cleanliness of the target photovoltaic component is lower than the cleanliness threshold and / or the current specific power generation efficiency is lower than the specific power generation efficiency threshold.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the photovoltaic component cleaning early warning method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the photovoltaic component cleaning early warning method according to any one of claims 1 to 7.

Images