Photovoltaic device automatic operation and maintenance management system and method based on computer vision

Abstract

The application discloses a photovoltaic device automatic operation and maintenance management system and method based on computer vision, relates to the technical field of photovoltaic device operation and maintenance management, and specifically comprises the following steps: acquiring the operation state information of each power generation component in a photovoltaic panel in a process in which the photovoltaic panel is affected by dynamic shading in real time, analyzing after acquisition, evaluating the power generation efficiency reduction degree of each power generation component in the photovoltaic panel caused by dynamic shading, and dividing each power generation component in the photovoltaic panel into low-impact components, medium-impact components and high-impact components; according to the division result of each power generation component in the photovoltaic panel, a power generation efficiency optimization mechanism is constructed, and different optimization measures are respectively taken for the low-impact components, the medium-impact components and the high-impact components. The application solves the problem of dynamic shading impact evaluation and optimization of photovoltaic devices, and realizes accurate improvement of power generation efficiency and stable adaptation of system operation.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic equipment operation and maintenance management technology, and specifically to an automated operation and maintenance management system and method for photovoltaic equipment based on computer vision. Background Technology

[0002] Photovoltaic (PV) equipment refers to devices that convert solar energy into electrical energy through photovoltaic (PV) panels, and is widely used in the field of solar power generation. The operation of PV equipment relies on multiple components, such as PV panels, inverters, and battery energy storage systems, and the stability of their performance directly affects power generation efficiency and energy utilization. With the continuous expansion of PV power generation, traditional manual operation and maintenance (O&M) models can no longer meet the demands for efficient and precise management. Manual inspections are not only time-consuming and labor-intensive, but also, due to the complexity of the environment, often fail to detect potential faults or problems in real time, leading to increased downtime and maintenance costs. Therefore, automated O&M management of PV equipment has become an inevitable trend. Automated O&M, by introducing computer vision technology, can monitor the operating status of equipment in real time and accurately locate equipment faults, such as stains, cracks, or shading, using image recognition algorithms, avoiding the limitations of manual inspection. Computer vision technology can not only perform all-weather, no-blind-spot equipment inspection, but also adapt to complex changes in lighting and environmental factors, significantly improving the accuracy and timeliness of fault diagnosis. Computer vision-based automated operation and maintenance management can significantly reduce labor costs, improve operation and maintenance efficiency, and ensure that photovoltaic equipment can operate efficiently and stably under various environmental conditions, thereby guaranteeing the continuous and stable output of photovoltaic power generation.

[0003] Existing computer vision-based automated operation and maintenance management technologies for photovoltaic (PV) equipment utilize high-resolution cameras or drones to capture real-time images of the PV equipment's operating status, transmitting these images to a backend system for processing. The backend system, relying on deep learning and image recognition algorithms, analyzes the PV panel surface in the images, automatically identifying potential fault characteristics such as stains, cracks, hot spots, and shading. It then generates fault diagnosis reports by labeling and classifying abnormal areas. Simultaneously, these systems can combine environmental monitoring data (such as light intensity, temperature, and wind speed) to comprehensively assess the PV equipment's operating environment and predict potential risks or trends in operational efficiency. Some advanced technologies also incorporate environmental adaptability correction algorithms, dynamically adjusting image illumination and enhancing contrast to improve fault identification accuracy under different weather and lighting conditions. Finally, the system pushes the analysis results to maintenance personnel through a cloud-based data analysis platform, providing specific maintenance suggestions or automatically generating repair work orders. This achieves efficient, accurate, and fully automated operation and maintenance management of PV equipment, significantly improving overall equipment operating efficiency and reliability.

[0004] The existing technology has the following shortcomings:

[0005] During the operation of photovoltaic (PV) equipment, when clouds move rapidly or obstacles in the environment (such as tree branches) sway, the individual power generation components in the PV panels, as the core power generation unit, are affected by dynamic shading. This results in uneven sunlight intensity and frequent changes in the area and duration of localized shading. This dynamic shading directly affects the power generation components in the PV panels, causing their power generation efficiency to fluctuate within a short period. Furthermore, the shading range and rate of change are difficult to predict due to complex environmental conditions, thus forming the dynamic shading characteristic. Since the multiple PV panel power generation components in a PV system are interconnected, the efficiency fluctuations of individual components further affect the overall operational balance of the equipment. However, existing computer vision-based automated operation and maintenance management technologies for PV equipment mainly rely on static image information, which can only roughly estimate the impact of shading on the overall PV system and lacks the ability to dynamically capture and accurately assess the declining power generation efficiency trends of individual power generation components within the PV panels. This technological limitation can lead to the failure to identify and optimize power generation components with high shading losses in a timely manner, making it impossible to take effective shading compensation measures. This not only reduces the overall power generation efficiency of photovoltaic equipment, but also exacerbates the aging and performance degradation of photovoltaic panels due to some power generation components being in a high shading state for a long time, thereby increasing the later operation and maintenance costs and repair difficulties.

[0006] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0007] The purpose of this invention is to provide an automated operation and maintenance management system and method for photovoltaic equipment based on computer vision, so as to solve the problems in the background art mentioned above.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a computer vision-based automated operation and maintenance management method for photovoltaic equipment, specifically including the following steps:

[0009] During the operation of photovoltaic equipment, computer vision technology is used to detect in real time whether dynamic shading occurs on the surface of the photovoltaic panel, which is the core power generation unit. When dynamic shading is detected on the surface of the photovoltaic panel, the individual power generation components in the photovoltaic panel used for power generation are identified.

[0010] Real-time acquisition of the operating status information of each power generation module in the photovoltaic panel during the process of the photovoltaic panel being affected by dynamic shading, and analysis after acquisition to assess the degree of power generation efficiency reduction of each power generation module in the photovoltaic panel caused by dynamic shading, and classifying each power generation module in the photovoltaic panel into low-impact modules, medium-impact modules and high-impact modules;

[0011] Based on the division of the various power generation components in the photovoltaic panel, a power generation efficiency optimization mechanism is constructed, and different optimization measures are implemented for low-impact, medium-impact, and high-impact components.

[0012] During the process of optimizing each power generation component in the photovoltaic panel by the power generation efficiency optimization mechanism, the optimization adjustment feedback information of each power generation component is obtained in real time, and after obtaining it, it is analyzed to evaluate whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component has achieved the expected results, and the power generation efficiency optimization mechanism is optimized based on the evaluation results.

[0013] The impact of dynamic shading on photovoltaic panels, the effectiveness of optimization measures, and the overall operation data of photovoltaic equipment are comprehensively monitored and analyzed to continuously improve the power generation efficiency optimization mechanism and enhance the operating efficiency and dynamic adaptability of photovoltaic equipment.

[0014] Preferably, the operating status information of each power generation module in the photovoltaic panel is acquired in real time during the process of the photovoltaic panel being affected by dynamic shading, and then analyzed to assess the degree of power generation efficiency reduction of each power generation module in the photovoltaic panel caused by dynamic shading, and the power generation modules in the photovoltaic panel are divided into low-impact modules, medium-impact modules, and high-impact modules. The specific steps include:

[0015] Real-time acquisition of the operating status information of each power generation component in the photovoltaic panel during the process of the photovoltaic panel being affected by dynamic shading, and preprocessing after acquisition;

[0016] Extract the light shading characteristic information and power generation fluctuation characteristic information from the preprocessed operating status information, and analyze them after extraction to generate the shading intensity index and power generation fluctuation coefficient of each power generation component.

[0017] A power generation efficiency degradation assessment model is constructed based on the shading intensity index and power generation fluctuation coefficient of each generated power generation component. The power generation assessment coefficient of each power generation component is generated by weighted summation and analyzed after generation to assess the degree of power generation efficiency degradation of each power generation component in the photovoltaic panel due to dynamic shading. The power generation components in the photovoltaic panel are then classified into low-impact components, medium-impact components, and high-impact components.

[0018] Preferably, the logic for obtaining the shading intensity index and power generation fluctuation coefficient of each power generation component is as follows:

[0019] Extract the shading characteristics from the preprocessed operational status information. Specifically, this includes the proportion of the shaded area on the surface of each power generation module to the total surface area at different times during a period of time when the photovoltaic panel is affected by dynamic shading, the light intensity received by each power generation module, and the rate of change of the light intensity received by each power generation module. Then, according to the time series, use functions to... , and To express, For a point in time, This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The percentage of the total surface area of ​​each power generation component that is shaded. This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The light intensity received by each power generation component This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The rate of change of solar intensity received by each power generation module is defined as the time period. , , It is a positive integer;

[0020] The solar irradiance received by each power generation component at different times during a period of time under dynamic shading is compared, and the maximum value is denoted as... ;

[0021] The shading intensity index for each power generation component is calculated using the following formula:

[0022]

[0023] In the formula, For the first Shading intensity index of each power generation component;

[0024] Extract the power generation fluctuation characteristics from the preprocessed operating status information. Specifically, this includes the actual power output value, the amplitude of change of the actual power output value, and the maximum power output value of each power generation module at different times during a period of time when the photovoltaic panel is affected by dynamic shading. Then, apply the actual power output value and the amplitude of change of the actual power output value of each power generation module at different times during a period of time when the photovoltaic panel is affected by dynamic shading to a function according to the time series. and To express, This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The actual power output value of each power generation component This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The variation in the actual power output value of each power generation component will be the first The maximum power output value of each power generation module is calibrated under standard test conditions as follows: ;

[0025] The power generation fluctuation coefficient of each power generation component is calculated using the following formula:

[0026]

[0027] In the formula, For the first The power generation fluctuation coefficient of each power generation component.

[0028] Preferably, the shading intensity index of each generated power generation component and power generation fluctuation coefficient A power generation efficiency degradation assessment model is constructed, and the power generation assessment coefficients for each power generation component are generated through weighted summation. According to the formula: ,in and The shading intensity index for each power generation component and power generation fluctuation coefficient The non-zero weight coefficients, and ;

[0029] The power generation evaluation coefficients of each generated power generation component Compared with the pre-set power generation assessment coefficient threshold range A comparison was conducted, and the degree of power generation efficiency reduction caused by dynamic shading in each photovoltaic module was assessed based on the comparison results. The photovoltaic modules were then categorized into low-impact, medium-impact, and high-impact modules. The specific comparison analysis and categorization are as follows:

[0030] like The power generation efficiency of this photovoltaic module is reduced to a low degree due to dynamic shading, and this module is classified as a low-impact module.

[0031] like The power generation efficiency of this photovoltaic module is reduced to a moderate degree due to dynamic shading, and this module is classified as a medium-impact module.

[0032] like The photovoltaic module is classified as a high-impact module because the power generation efficiency of this module is reduced to a high degree due to dynamic shading.

[0033] Preferably, a power generation efficiency optimization mechanism is constructed based on the division results of each power generation component in the photovoltaic panel. Specifically, different optimization adjustment parameters are set according to the division results of low-impact components, medium-impact components, and high-impact components to form a power generation efficiency optimization mechanism. This optimization mechanism is based on the dynamic feedback of the power generation evaluation coefficient of each power generation component, as well as the shading intensity index and power generation fluctuation coefficient. Through pre-set optimization rules, it automatically determines the adjustment method and adjustment range of the component's operating parameters.

[0034] Different optimization measures are applied to low-impact, medium-impact, and high-impact components, as follows:

[0035] The optimization measures for low-impact components are as follows: the run-hold parameters in the optimization mechanism are used to maintain the current operating state without adjusting the parameters;

[0036] The optimization measures implemented for the affected components are as follows: using the operation adjustment parameters in the optimization mechanism, the operation parameters of the power generation components are clearly adjusted, including resetting the current and voltage characteristic values ​​and implementing the shading compensation strategy to reduce the impact of dynamic shading on power generation performance.

[0037] The optimization measures for high-impact components are as follows: using the operation optimization parameters in the optimization mechanism, the operation mode of the power generation components is adjusted, including dynamically changing the distribution of the shaded area, redistributing the operating load of the components, and implementing full-coverage compensation measures for the shaded area to ensure the restoration of power generation performance.

[0038] Preferably, during the process of optimizing each power generation component in the photovoltaic panel by the power generation efficiency optimization mechanism, the optimization adjustment feedback information of each power generation component is acquired in real time, analyzed after acquisition, and the optimization effect of the power generation efficiency optimization mechanism on each power generation component is evaluated to determine whether it meets expectations. Based on the evaluation results, the power generation efficiency optimization mechanism is then optimized. Specifically, this includes the following steps:

[0039] During the process of optimizing each power generation component in the photovoltaic panel by the power generation efficiency optimization mechanism, the optimization adjustment feedback information of each power generation component is obtained in real time and preprocessed after acquisition;

[0040] The optimized output characteristic information and optimized response characteristic information are extracted from the optimized adjustment feedback information of each preprocessed power generation component, and analyzed after extraction to generate the optimized output consistency index and optimized response sensitivity coefficient of each power generation component.

[0041] An optimization effect evaluation model is constructed based on the optimized output consistency index and optimized response sensitivity coefficient of each generated power generation component. The optimization evaluation coefficient of each power generation component is generated by weighted summation and analyzed after generation to evaluate whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component has reached the expectation. Based on the evaluation results, the power generation efficiency optimization mechanism is optimized.

[0042] Preferably, the logic for obtaining the optimized output consistency index and optimized response sensitivity coefficient of each power generation component is as follows:

[0043] The optimized output characteristic information is extracted from the preprocessed optimization adjustment feedback information of each power generation component. Specifically, this includes the actual output power of each power generation component at different times during a certain period of the optimization process, the magnitude of change in actual output power, and the target power value set for each power generation component by the power generation efficiency optimization mechanism. These are then calibrated as follows: , and , This indicates a period of time during the optimization process. Time of the first The actual output power of each power generation component This indicates a period of time during the optimization process. Time of the first The variation range of the actual output power of each power generation component The power generation efficiency optimization mechanism is represented as the first... The target power value set for each power generation component. , , and All are positive integers;

[0044] The optimal output consistency index of each power generation component is calculated using the following formula:

[0045]

[0046] In the formula, For the first The optimized output consistency index of each power generation component;

[0047] The optimization response characteristic information is extracted from the preprocessed optimization adjustment feedback information of each power generation component. Specifically, this includes the percentage change in the ratio of actual input energy to output energy of each power generation component at different times during the optimization process, the rate of change in the power generation efficiency of each power generation component, and the time interval from receiving the optimization command to achieving the adjustment effect for each power generation component. These are then calibrated as follows: , and , This indicates a period of time during the optimization process. Time of the first The percentage change in the ratio of actual input energy to output energy of each power generation component. This indicates a period of time during the optimization process. Time of the first The rate of change in the power generation efficiency of each power generation component Indicates the first step in the optimization process. The time interval between receiving an optimization command and achieving the desired adjustment effect for each power generation component;

[0048] The optimized response sensitivity coefficient of each power generation component is calculated using the following formula:

[0049]

[0050] In the formula, For the first Optimized response sensitivity coefficients for each power generation component.

[0051] Preferably, the optimized output consistency index of each generated power generation component and optimize response sensitivity coefficient An optimization performance evaluation model is constructed, and optimization evaluation coefficients for each power generation component are generated through weighted summation. According to the formula: ,in and The optimized output consistency index for each power generation component and optimize response sensitivity coefficient The non-zero weight coefficients, and ;

[0052] The optimized evaluation coefficients of each generated power generation component Compared with the pre-set optimization evaluation coefficient thresholds for each power generation component A comparison was conducted, and the optimization effect of the power generation efficiency optimization mechanism on each power generation component was evaluated based on the comparison results. The power generation efficiency optimization mechanism was then optimized based on the evaluation results. The specific comparative analysis is as follows:

[0053] like The power generation efficiency optimization mechanism has achieved the expected optimization effect on the power generation component, and there is no need to optimize the power generation efficiency optimization mechanism.

[0054] like The power generation efficiency optimization mechanism has not achieved the expected optimization effect on the power generation component, and the mechanism needs to be optimized. Specifically, this includes: readjusting the operating parameter settings in the optimization strategy; improving the feedback information processing mechanism to enhance the real-time performance and accuracy of optimization adjustments; dynamically updating the optimization effect evaluation model to adapt it to current conditions based on new operating data; and optimizing the execution process to ensure that the power generation efficiency optimization mechanism can improve the power generation performance of each power generation component by reducing response latency and improving the efficiency of optimization command execution.

[0055] Preferably, the computer vision-based automated operation and maintenance management system for photovoltaic equipment includes a dynamic shading detection module, an operation status assessment module, a power generation efficiency optimization module, an optimization effect feedback module, and a comprehensive monitoring and improvement module.

[0056] The dynamic shading detection module uses computer vision technology to detect in real time whether dynamic shading occurs on the surface of the photovoltaic panel, which is the core power generation unit, during the operation of the photovoltaic equipment. When dynamic shading is detected on the surface of the photovoltaic panel, the module identifies the various power generation components in the photovoltaic panel that are used for power generation.

[0057] The operation status assessment module acquires the operation status information of each power generation component in the photovoltaic panel in real time when the photovoltaic panel is affected by dynamic shading, and analyzes the information to assess the degree of power generation efficiency reduction caused by dynamic shading of each power generation component in the photovoltaic panel, and classifies each power generation component in the photovoltaic panel into low-impact components, medium-impact components and high-impact components.

[0058] The power generation efficiency optimization module constructs a power generation efficiency optimization mechanism based on the division of each power generation component in the photovoltaic panel, and implements different optimization measures for low-impact, medium-impact, and high-impact components.

[0059] The optimization effect feedback module acquires the optimization adjustment feedback information of each power generation component in the photovoltaic panel in real time during the process of the power generation efficiency optimization mechanism optimizing each power generation component. After acquisition, it analyzes and evaluates whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component has met expectations, and optimizes the power generation efficiency optimization mechanism based on the evaluation results.

[0060] The comprehensive monitoring and improvement module comprehensively monitors and analyzes the impact of dynamic shading on photovoltaic panels, the effectiveness of optimization measures, and the overall operation data of photovoltaic equipment. It continuously improves the power generation efficiency optimization mechanism and enhances the operating efficiency and dynamic adaptability of photovoltaic equipment.

[0061] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0062] 1. This invention utilizes computer vision technology to achieve real-time detection of dynamic shading in photovoltaic (PV) equipment and accurate identification of the status of power generation components, significantly improving the ability to capture and analyze dynamic shading. By combining the acquisition and modeling of the shading intensity index and power generation fluctuation coefficient, the impact of dynamic shading on the power generation efficiency of each component can be accurately assessed, classifying components into different levels of impact. Differentiated optimization measures, such as adjusting operating parameters and shading compensation, are implemented for low-impact, medium-impact, and high-impact components, further ensuring the improvement of the overall power generation performance of the PV panels. Simultaneously, during the optimization process, the system dynamically analyzes the feedback information from optimization adjustments, evaluates and optimizes the power generation efficiency optimization mechanism in real time, achieving closed-loop management of the PV equipment's operating status. This multi-stage linkage technology provides stable and efficient operational assurance for PV equipment in complex dynamic environments.

[0063] 2. This invention employs a calculation model combining shading intensity index and power generation fluctuation coefficient to comprehensively quantify the impact of dynamic shading on power generation components, providing a scientific basis for classification and optimization. Another advantage is its intelligence. Through the automated adjustment of the power generation efficiency optimization mechanism and the dynamic updating of the optimization effect evaluation model, photovoltaic equipment can continuously optimize its operating strategy based on real-time operating data, enhancing the system's autonomous adaptability and decision-making capabilities.

[0064] 3. This invention covers the entire process from shading detection to optimization and adjustment, forming a closed-loop management system from problem identification to resolution. Differentiated optimization measures effectively improve overall power generation efficiency, reduce equipment aging and maintenance costs, and enhance the system's adaptability in dynamic environments. Compared to existing technologies, this solution not only improves performance but also achieves superior economic benefits. Attached Figure Description

[0065] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0066] Figure 1 This is a flowchart illustrating the automated operation and maintenance management system and method for photovoltaic equipment based on computer vision, as described in this invention.

[0067] Figure 2 This is a schematic diagram of the modules of the computer vision-based automated operation and maintenance management system and method for photovoltaic equipment of the present invention. Detailed Implementation

[0068] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0069] This invention provides, for example Figure 1 The computer vision-based automated operation and maintenance management method for photovoltaic equipment, as shown, specifically includes the following steps:

[0070] During the operation of photovoltaic equipment, computer vision technology is used to detect in real time whether dynamic shading occurs on the surface of the photovoltaic panel, which is the core power generation unit. When dynamic shading is detected on the surface of the photovoltaic panel, the individual power generation components in the photovoltaic panel used for power generation are identified.

[0071] Real-time detection of dynamic shading on photovoltaic (PV) panel surfaces using computer vision technology includes: deploying high-precision cameras in the PV installation area for real-time image acquisition; preprocessing the images such as noise reduction, color correction, and edge enhancement; using convolutional neural network image segmentation algorithms to divide the image into normal power generation areas and suspected shading areas; and further analyzing the presence of shading by considering brightness changes, texture features, and boundary characteristics. Simultaneously, temporal analysis of consecutive frames is performed to extract shading trajectories and durations, comprehensively capturing the range, intensity, and changing characteristics of dynamic shading. Dynamic shading mainly refers to shading phenomena on PV panel surfaces caused by cloud movement, tree branch swaying, etc., characterized by constantly changing shading range, intensity, and duration, significantly impacting power generation efficiency.

[0072] After detecting dynamic shading, the components used for power generation within the photovoltaic panel and their shading status are further identified. A region localization algorithm is used to match the photovoltaic panel surface image with a partitioned model of the power generation components, accurately identifying the corresponding region of each component in the image. The analysis results are then validated using operational data such as light intensity distribution and temperature changes. Simultaneously, through cross-analysis of the dynamic shading trajectory and component position, the shaded components and their degree of shading are located, thus eliminating interference from non-power generation areas such as the frame or support structure, ensuring the accuracy of the identification.

[0073] Accurate identification of shaded power generation modules is crucial for ensuring the efficient operation of photovoltaic (PV) equipment. Dynamic shading leads to a decrease in module efficiency, affecting not only individual modules but also potentially impacting other modules, reducing overall power generation efficiency and stability. Failure to accurately identify shaded modules makes it difficult to assess their performance and implement targeted optimizations, potentially resulting in uncompensated shading losses, increased equipment aging, and higher maintenance costs. This technical solution combines computer vision and module identification technologies to provide high-quality data support for subsequent evaluation and optimization, ensuring stable and efficient equipment operation even in dynamic environments.

[0074] Real-time acquisition of the operating status information of each power generation module in the photovoltaic panel during the process of the photovoltaic panel being affected by dynamic shading, and analysis after acquisition to assess the degree of power generation efficiency reduction of each power generation module in the photovoltaic panel caused by dynamic shading, and classifying each power generation module in the photovoltaic panel into low-impact modules, medium-impact modules and high-impact modules;

[0075] In this embodiment, the operating status information of each power generation module in the photovoltaic panel is acquired in real time during the process of the photovoltaic panel being affected by dynamic shading. After acquisition, the information is analyzed to evaluate the degree of power generation efficiency reduction of each power generation module in the photovoltaic panel caused by dynamic shading. The power generation modules in the photovoltaic panel are then divided into low-impact modules, medium-impact modules, and high-impact modules. The specific steps include:

[0076] Real-time acquisition of the operating status information of each power generation component in the photovoltaic panel during the process of the photovoltaic panel being affected by dynamic shading, and preprocessing after acquisition;

[0077] Real-time acquisition of operational status information for each power generation module within a photovoltaic (PV) panel under dynamic shading can be achieved by combining multiple sensors on the PV equipment with computer vision technology. Specifically, a high-precision camera is first deployed on the PV panel surface to continuously acquire surface images and capture image features of dynamic shading. Simultaneously, a light intensity sensor and a power monitoring module record the light intensity and actual power generation data of each power generation module in real time. This data has time-series characteristics, reflecting the changes in dynamic shading and power fluctuations. The image data acquired by the camera is processed in real time using computer vision algorithms to segment the PV panel into regions for each power generation module, thereby identifying and labeling the light conditions and shading areas of each module. Simultaneously, the data from the light intensity sensor is combined with the image processing results to mark the light level of each power generation module; the data from the power monitoring module is used to acquire information on changes in power generation. Through these methods, the light conditions, shading status, and power generation fluctuation information of each power generation module are integrated into operational status information and updated in real time to adapt to dynamic environmental changes.

[0078] After acquisition, operational status information needs to be preprocessed to ensure the accuracy and efficiency of subsequent analysis. The main goal of preprocessing is to eliminate noise and redundant information in the data, improve data quality, and standardize the data format for unified analysis. Specific preprocessing operations include the following: First, image data preprocessing: noise interference in the image is eliminated using denoising algorithms (such as Gaussian filtering), the contours of the power generation component area in the image are enhanced using edge detection algorithms, and feature information of the power generation component area is extracted using image segmentation algorithms; Second, illumination intensity data preprocessing: the sliding window averaging method is used to smooth fluctuations in illumination intensity and eliminate errors caused by short-term random changes; Third, power generation data preprocessing: anomaly detection (such as removing outliers that significantly exceed the normal range) and interpolation completion (such as using linear interpolation algorithms to complete missing data in cases of excessively large data acquisition intervals). Through these preprocessing steps, operational status information can be transformed into high-quality, standardized analytical input, thereby providing reliable data support for subsequent characteristic information extraction and parameter calculation.

[0079] Extract the light shading characteristic information and power generation fluctuation characteristic information from the preprocessed operating status information, and analyze them after extraction to generate the shading intensity index and power generation fluctuation coefficient of each power generation component.

[0080] Extracting shading characteristics and power generation fluctuation characteristics from preprocessed operational status information can be achieved through a combination of data analysis and computer vision techniques. First, for shading characteristics, the photovoltaic panel surface is divided into specific regions for each power generation component from the preprocessed image data using region segmentation algorithms in computer vision (such as convolutional neural network-based segmentation models), and the shading region boundaries of each component are identified. Further, the shading area ratio is obtained by calculating the pixel ratio of the shading region. Simultaneously, data from a light intensity sensor is combined to analyze the light intensity level and its relative change in the shading region, and the rate of change of light intensity over a period of time is calculated. This allows for the extraction of shading characteristics, including the shading area ratio, light intensity distribution, and rate of change of light intensity. Secondly, regarding the power generation fluctuation characteristics, the power fluctuation amplitude is calculated from the time-series data of the power generation monitoring module using statistical and signal processing algorithms. For example, the principal components of the power change frequency are extracted using Fast Fourier Transform, or the standard deviation of the power fluctuation is calculated after smoothing noise using moving average filtering. Simultaneously, the deviation ratio of the real-time power is calculated based on the reference maximum power, and the above results are combined to generate key data reflecting the power generation fluctuation characteristics. This information extraction method efficiently separates two types of characteristic information from the operating status information, providing accurate input for subsequent parameter calculations.

[0081] A power generation efficiency degradation assessment model is constructed based on the shading intensity index and power generation fluctuation coefficient of each generated power generation component. The power generation assessment coefficient of each power generation component is generated by weighted summation and analyzed after generation to assess the degree of power generation efficiency degradation of each power generation component in the photovoltaic panel due to dynamic shading. The power generation components in the photovoltaic panel are then classified into low-impact components, medium-impact components, and high-impact components.

[0082] In this embodiment, the logic for obtaining the shading intensity index and power generation fluctuation coefficient of each power generation component is as follows:

[0083] Extract the shading characteristics from the preprocessed operational status information. Specifically, this includes the proportion of the shaded area on the surface of each power generation module to the total surface area at different times during a period of time when the photovoltaic panel is affected by dynamic shading, the light intensity received by each power generation module, and the rate of change of the light intensity received by each power generation module. Then, according to the time series, use functions to... , and To express, For a point in time, This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The percentage of the total surface area of ​​each power generation component that is shaded. This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The light intensity received by each power generation component This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The rate of change of solar intensity received by each power generation module is defined as the time period. , , It is a positive integer;

[0084] To acquire three types of data in real time—the proportion of the surface area of ​​each power generation module that is shaded relative to the total surface area, the light intensity received by each power generation module, and the rate of change of the light intensity received by each power generation module—can be achieved by combining computer vision technology with light sensors. Specifically, firstly, a high-precision camera installed on the photovoltaic equipment continuously acquires images of the photovoltaic panel surface. Using image segmentation technology in computer vision, the photovoltaic panel surface is divided into independent regions for each power generation module, and the shaded portion of each module's surface is detected. By calculating the ratio of the pixel area of ​​the shaded portion to the total pixel area of ​​the power generation module, the proportion of the shaded area on each power generation module's surface is determined in real time. Secondly, a light intensity sensor installed on each power generation module records the light intensity data received by the current module in real time. This data is aggregated through a sensor network and sent to a central processing system for analysis. Finally, the rate of change of light intensity is obtained by analyzing the differences in light intensity data between adjacent time points. The system processes multiple consecutively sampled light intensity data, extracts the trend of light intensity change over time, and generates the rate of change of light intensity for each power generation module. This combination of cameras and light sensors ensures comprehensive dynamic monitoring of shading and lighting conditions on the photovoltaic panel surface, providing accurate basic data for subsequent shading intensity analysis.

[0085] The solar irradiance received by each power generation component at different times during a period of time under dynamic shading is compared, and the maximum value is denoted as... ;

[0086] The shading intensity index for each power generation component is calculated using the following formula:

[0087]

[0088] In the formula, For the first Shading intensity index of each power generation component;

[0089] Calculate the shading intensity index of each power generation component. This formula aims to comprehensively quantify the impact of dynamic shading on the illumination conditions of individual power generation modules and provide a reliable numerical basis for subsequent power generation efficiency assessments. It captures the spatial, intensity, and dynamic characteristics of shading through multi-level computational steps. Specifically, the integral operation is used to accumulate the shading impact over a given time period to reflect the dynamic characteristics and temporal distribution of shading; the shading area ratio... It describes the spatial extent of the photovoltaic panel surface that is shaded, reflecting the degree of spatial impact of the shading; It is a non-linear function that uses exponential operations to amplify the normalized value of light intensity, ensuring that the effects of shading are significantly highlighted under low light conditions, while avoiding the imbalance caused by direct linear accumulation under high light intensity; the rate of change of light intensity. The dynamic fluctuations in light intensity were captured, reflecting the dynamic characteristics of shading affecting photovoltaic modules in real time. The overall calculation results, by logarithmically processing the average over a time period, smooth out situations where the shading impact is weak, while simultaneously causing the results to rise rapidly when shading is significant, emphasizing the severity of the shading. This multi-layered calculation logic enables the shading intensity index to comprehensively reflect the dynamic characteristics, cumulative effects over time, and spatial range of shading, exhibiting strong adaptability and accuracy, and providing a scientific basis for subsequent analysis.

[0090] No. Shading intensity index of individual power generation components The magnitude of the light intensity directly reflects the overall impact of sunlight on the component under dynamic shading conditions, and is positively correlated with the degree of decline in its power generation efficiency. When A larger value indicates a higher proportion of shaded area, lower light intensity, or a larger rate of change in light intensity. These factors all lead to a significant decrease in the power generation efficiency of the module; conversely, when... A lower value indicates that the component is less affected by shading, and the power generation efficiency is less affected. Therefore, by evaluating... The magnitude of the shading can precisely quantify the dynamic impact of shading on the power generation efficiency of each power generation component, providing a core basis for subsequent efficiency assessment and component classification. For example, in the assessment model of power generation efficiency decline, The value can be used as a weighting factor and combined with other characteristic data (such as power fluctuation coefficient) to generate power generation assessment coefficient, thereby further accurately assessing the degree of efficiency reduction caused by shading and providing scientific guidance for subsequent optimization measures.

[0091] Extract the power generation fluctuation characteristics from the preprocessed operating status information. Specifically, this includes the actual power output value, the amplitude of change of the actual power output value, and the maximum power output value of each power generation module at different times during a period of time when the photovoltaic panel is affected by dynamic shading. Then, apply the actual power output value and the amplitude of change of the actual power output value of each power generation module at different times during a period of time when the photovoltaic panel is affected by dynamic shading to a function according to the time series. and To express, This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The actual power output value of each power generation component This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The variation in the actual power output value of each power generation component will be the first The maximum power output value of each power generation module is calibrated under standard test conditions as follows: ;

[0092] To acquire three types of data in real time—the actual power output value of each photovoltaic module, the magnitude of change in the actual power output value, and the maximum power output value of each photovoltaic module under standard test conditions—can be achieved by combining power sensors and a data processing system. First, high-precision power sensors are installed at the output end of each photovoltaic module. These sensors monitor the actual power output value of the module in real time and record the data in a time-series format, forming a dynamic data stream reflecting changes in module power. Second, the magnitude of change in the actual power output value is obtained through differential analysis of the data collected by the power sensors. Specifically, the software system compares the actual power output values ​​at different times, calculates the power difference between adjacent time points, and extracts the magnitude of change. This process, combined with dynamic time window settings, ensures that the calculation of the magnitude of change captures the trend of the impact of shading on power output. Finally, the maximum power output value of each photovoltaic module under standard test conditions can be directly obtained through preset values. Data for standard test conditions is typically provided by the photovoltaic module manufacturer and stored in the equipment's operational database. By correlating the real-time collected actual power output values ​​and their variation amplitudes with the standard maximum power output values, the software system can dynamically monitor and analyze the power fluctuation characteristics of each power generation component, providing high-quality data support for subsequent analysis and evaluation.

[0093] The power generation fluctuation coefficient of each power generation component is calculated using the following formula:

[0094]

[0095] In the formula, For the first The power generation fluctuation coefficient of each power generation component.

[0096] Calculate the power generation fluctuation coefficient of each power generation component. This is to comprehensively quantify the instability of power output under dynamic shading conditions and its impact on power generation performance. The settings for each calculation step in the formula are designed to highlight the contribution of different characteristics to power fluctuations. First, exponential calculation is used. To describe the degree of insufficiency of the actual power output value relative to the maximum power, this part amplifies the impact of low power output through nonlinear attenuation, making the period of significant power deficiency contribute more to the overall coefficient; secondly, logarithmic operations are used. The system enhances sensitivity to power fluctuations, smoothing out the impact of small fluctuations while emphasizing the contribution of large power fluctuations to the overall coefficient. These two parts, through summation, express the combined effect of power insufficiency and fluctuations, ensuring a comprehensive reflection of the strength of the shading impact. Furthermore, by integrating power fluctuations over the entire time period, the system captures the time effect of dynamic changes, and normalizes the impact through averaging, making the results more universal and comparable. Finally, square root calculation balances the dimensional differences between different characteristics, ensuring greater interpretability. This calculation logic accurately reflects the complex impact of dynamic shading on the power fluctuations of power generation components, providing a scientific basis for efficiency assessment.

[0097] No. Power generation fluctuation coefficient of individual power generation components The size of the value is closely related to the assessment of the degree of power generation efficiency reduction caused by dynamic shading of the component. The larger the value, the more unstable the power output of the component under dynamic shading conditions, manifested as a higher proportion of insufficient power time or larger power fluctuations. These factors directly reflect the negative impact of shading on power generation efficiency. When the power fluctuation coefficient is high, it indicates that shading not only reduces the average power output of the component but also significantly increases the instability of power generation, leading to a more severe efficiency decline; conversely, when... A smaller value indicates that the power output of the component is relatively stable, and the impact of shading on power generation efficiency is minimal. Therefore, the magnitude of the power generation fluctuation coefficient can serve as an important indicator for measuring the degree of impact of dynamic shading on power generation components, providing a key basis for subsequent efficiency degradation assessments, and working in conjunction with other assessment parameters to generate comprehensive assessment results, supporting optimization and adjustment strategies for power generation components.

[0098] In this embodiment, the shading intensity index of each generated power generation component is... and power generation fluctuation coefficient A power generation efficiency degradation assessment model is constructed, and the power generation assessment coefficients for each power generation component are generated through weighted summation. According to the formula: ,in and The shading intensity index for each power generation component and power generation fluctuation coefficient The non-zero weight coefficients, and ;

[0099] The power generation efficiency degradation assessment model is constructed, and the power generation assessment coefficients for each power generation component are generated through weighted summation. This is achieved by combining two core parameters: the shading intensity index and the power generation fluctuation coefficient. First, the shading intensity index and power generation fluctuation coefficient are extracted from each power generation component. These two parameters quantify the impact of dynamic shading on illumination conditions and power generation performance stability, respectively. Then, based on the actual application scenarios and technical requirements, appropriate weighting coefficients are assigned to these two parameters. and The weighting coefficients are determined through experimental data or model training. The value of is usually related to the relative contribution of the shading intensity index to the decrease in power generation efficiency, while This is related to the degree of impact of power fluctuations on power generation performance. This weighting can be dynamically adjusted according to different environmental conditions (such as fluctuations in sunlight intensity and frequency of shading) to ensure the model's adaptability and accuracy. In the specific calculation, the software system multiplies the shading intensity index of each component by its corresponding weighting coefficient. Power generation fluctuation coefficient multiplied by weight The weighted values ​​of the two factors are then summed to generate the power generation assessment coefficient for the component. This process is achieved through real-time data input and algorithm processing, constructing a quantitative assessment model for the combined impact of dynamic shading and power fluctuations, providing a scientific basis for subsequent efficiency degradation analysis and component classification.

[0100] The power generation evaluation coefficients of each generated power generation component Compared with the pre-set power generation assessment coefficient threshold range A comparison was conducted, and the degree of power generation efficiency reduction caused by dynamic shading in each photovoltaic module was assessed based on the comparison results. The photovoltaic modules were then categorized into low-impact, medium-impact, and high-impact modules. The specific comparison analysis and categorization are as follows:

[0101] like The power generation efficiency of this photovoltaic module is reduced to a low degree due to dynamic shading, and this module is classified as a low-impact module.

[0102] This indicates that the module's power generation efficiency degradation due to dynamic shading is relatively low. In other words, the shading area of ​​the module during photovoltaic panel operation is small, the light intensity remains relatively stable, and the fluctuation range of power generation is within an acceptable range. This means that the module's power generation performance is largely unaffected by dynamic shading, and its output power can stably approach the maximum power output under standard test conditions. The impact is that the module does not require additional optimization measures and can maintain its existing operating state to conserve system resources. This allows more priority to be given to adjustments and optimizations for other more severely affected modules, thereby maximizing the overall efficiency of the photovoltaic system.

[0103] like The power generation efficiency of this photovoltaic module is reduced to a moderate degree due to dynamic shading, and this module is classified as a medium-impact module.

[0104] This indicates that the reduction in power generation efficiency caused by dynamic shading is moderate. At this point, the shading area ratio, changes in irradiance, and power fluctuation characteristics are all within a certain range, having some impact on power generation performance, but not to a severe degree. This means that the output power of the module is significantly affected by dynamic shading, but remains within an adjustable and optimizable range. The impact is that the module may require moderate optimization measures, such as adjusting the distribution of the shading area or making small-scale optimizations to the module's operating parameters to reduce the impact of dynamic shading while avoiding unnecessary resource consumption from excessive intervention. By making appropriate adjustments to the moderately affected module, the overall power generation efficiency of the photovoltaic equipment can be further improved.

[0105] like The photovoltaic module is classified as a high-impact module because the power generation efficiency of this module is reduced to a high degree due to dynamic shading.

[0106] This situation indicates a severe decline in the module's power generation efficiency due to dynamic shading. In this case, the module may have been in a large shading area for an extended period, resulting in a significant decrease in sunlight intensity and markedly unstable power output fluctuations, far below the maximum power output under standard testing conditions. This means the module has a significant negative impact on the overall operation of the photovoltaic system, and without optimization measures, it could lead to further declines in power generation efficiency and affect the overall balanced operation of the system. Therefore, this module requires priority optimization measures with strong intervention, such as rapidly adjusting the shading area distribution, reallocating operating parameters, and even physically intervening in the shading source if necessary. This prioritized optimization can not only restore the power generation performance of the high-impact module but also reduce its negative ripple effects on the entire photovoltaic system, improving the overall power generation efficiency and stability of the system.

[0107] The pre-defined threshold range for power generation assessment coefficients can be determined through a combination of historical data analysis and experimental calibration. Specifically, this involves collecting long-term data on photovoltaic equipment under different environments (such as shading intensity index, power generation fluctuation coefficient, and actual power generation efficiency); using cluster analysis and statistical modeling to identify characteristic patterns of the impact of dynamic shading on power generation efficiency; establishing a benchmark model; classifying the impact into three categories: low, medium, and high; and extracting corresponding typical value ranges. Experimental calibration is then conducted by simulating different shading conditions, dynamically adjusting the upper and lower limits of the thresholds, and using a multi-objective optimization algorithm to balance the classification effect. This process can be automated by a software system, providing a scientific basis for efficiency assessment and classification.

[0108] Based on the division of the various power generation components in the photovoltaic panel, a power generation efficiency optimization mechanism is constructed, and different optimization measures are implemented for low-impact, medium-impact, and high-impact components.

[0109] In this embodiment, a power generation efficiency optimization mechanism is constructed based on the division results of each power generation component in the photovoltaic panel. Specifically, different optimization adjustment parameters are set according to the division results of low-impact components, medium-impact components, and high-impact components to form a power generation efficiency optimization mechanism. This optimization mechanism is based on the dynamic feedback of the power generation evaluation coefficient of each power generation component, as well as the shading intensity index and power generation fluctuation coefficient. Through the pre-set optimization rules, it automatically determines the adjustment method and adjustment range of the component's operating parameters.

[0110] To achieve the goal of "constructing a power generation efficiency optimization mechanism based on the classification of various power generation components in a photovoltaic panel," a dynamic parameter adjustment model combined with data feedback analysis can be used. Specifically, the classification results of low-impact, medium-impact, and high-impact components need to be input into the optimization mechanism, which consists of a rule base based on power generation assessment coefficients and a dynamic adjustment model. The rule base pre-defines operational adjustment strategies for different categories of components, including specific measures such as maintaining status, optimizing parameters, and prioritizing compensation. The optimization mechanism first uses the power generation assessment coefficients and real-time updated shading intensity index and power generation fluctuation coefficient data to dynamically analyze the component operating status and calculates the adjustment priority of each component using a multi-objective optimization algorithm (such as a genetic algorithm or fuzzy logic model). Subsequently, according to the optimization rules in the rule base, optimization adjustment parameters are assigned to each component, such as current and voltage characteristic values, shading compensation strategies, and operating load allocation. The setting of these parameters comprehensively considers the current status of the components, historical operating data, and dynamic feedback information through the optimization algorithm, ensuring that the adjustment measures are scientific and adaptable. This approach is chosen because photovoltaic (PV) equipment is significantly affected by environmental factors during actual operation. The efficiency decline caused by dynamic shading necessitates precise adjustments to the operating parameters of different components to optimize overall power generation performance. The optimization mechanism utilizes real-time feedback data to quickly identify and classify components affected to varying degrees, and to develop targeted, differentiated adjustment strategies. This avoids resource waste or unnecessary adjustments, thereby improving the overall operating efficiency and stability of the equipment. Furthermore, this software-based dynamic optimization method adapts to different operating environments, offering strong flexibility and scalability, and providing core technological support for intelligent operation and maintenance of PV equipment.

[0111] Different optimization measures are applied to low-impact, medium-impact, and high-impact components, as follows:

[0112] The optimization measures for low-impact components are as follows: the run-hold parameters in the optimization mechanism are used to maintain the current operating state without adjusting the parameters;

[0113] Optimization measures for low-impact modules can be implemented through a maintenance mechanism within the software system. Specifically, based on the comparison results of the module's power generation assessment coefficient, the system sets the power generation units assessed as low-impact modules to "maintain mode." In maintain mode, the system monitors the module's power generation parameters in real time to ensure that the module's current operating state remains within a stable range; for example, key parameters such as current, voltage, and power output are all within normal ranges. In this mode, the optimization mechanism suspends adjustment commands for the module, retaining only basic operational monitoring functions to reduce unnecessary computation and resource consumption. The significance of this approach is that excessive intervention for modules whose power generation performance is not significantly affected by dynamic shading may lead to resource waste and even disrupt the overall operational balance. Therefore, by maintaining the current operating state, more computational and optimization resources can be allocated to medium-impact and high-impact modules, thereby improving the optimization efficiency of the entire photovoltaic system.

[0114] The optimization measures implemented for the affected components are as follows: using the operation adjustment parameters in the optimization mechanism, the operation parameters of the power generation components are clearly adjusted, including resetting the current and voltage characteristic values ​​and implementing the shading compensation strategy to reduce the impact of dynamic shading on power generation performance.

[0115] Optimization measures for moderately affected modules are implemented through an operational adjustment mechanism. This mechanism dynamically generates adjustment parameters based on the module's power generation assessment coefficient, shading intensity index, and power generation fluctuation coefficient. The specific implementation involves two parts: First, resetting the current and voltage characteristic values. The system analyzes the module's real-time operating point, combines it with shading intensity data, calculates a new maximum power point, and adjusts the inverter's operating parameters to make the module operate at the new optimal power point. Second, implementing shading compensation strategies, such as using dynamic distributed maximum power point tracking (MPPT) to optimize the output efficiency of the power generation module under partial shading conditions, while simultaneously adjusting local power allocation based on dynamic data of the shading area. The significance of this optimization method is that although moderately affected modules are significantly impacted by shading, by appropriately adjusting operating parameters, the module's power generation performance can be significantly improved, avoiding greater efficiency losses, thereby maximizing the module's power generation capacity while ensuring operational stability.

[0116] The optimization measures for high-impact components are as follows: using the operation optimization parameters in the optimization mechanism, the operation mode of the power generation components is adjusted, including dynamically changing the distribution of the shaded area, redistributing the operating load of the components, and implementing full-coverage compensation measures for the shaded area to ensure the restoration of power generation performance.

[0117] Optimization measures for high-impact components are implemented through an operational optimization mechanism, employing a multi-level adjustment approach for focused optimization. Specifically, this involves: First, dynamically analyzing the distribution characteristics of shading areas using computer vision data, and adjusting the arrangement of shading areas via software commands, such as controlling the action of adjustable shading devices or notifying physical maintenance personnel to remove shading sources; second, redistributing the operating load of components, distributing the load of severely shaded components to other components to balance overall output power; and finally, implementing full-coverage compensation measures for shading areas, such as activating backup power generation components or adjusting the power output of the energy storage system to compensate for the power shortfall caused by shading. The significance of this optimization approach lies in the fact that if high-impact components are not optimized in a timely manner, they will have a significant negative impact on the operating efficiency and stability of the entire photovoltaic system. Therefore, through multi-level optimization, power generation performance can be restored to the maximum extent possible, while reducing the cascading impact on other components and equipment, ensuring the overall high efficiency and reliability of the system.

[0118] During the process of optimizing each power generation component in the photovoltaic panel by the power generation efficiency optimization mechanism, the optimization adjustment feedback information of each power generation component is obtained in real time, and after obtaining it, it is analyzed to evaluate whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component has achieved the expected results, and the power generation efficiency optimization mechanism is optimized based on the evaluation results.

[0119] In this embodiment, during the process of optimizing each power generation component in the photovoltaic panel by the power generation efficiency optimization mechanism, the optimization adjustment feedback information of each power generation component is acquired in real time, analyzed after acquisition, and the optimization effect of the power generation efficiency optimization mechanism on each power generation component is evaluated to determine whether it meets expectations. Based on the evaluation results, the power generation efficiency optimization mechanism is optimized, specifically including the following steps:

[0120] During the process of optimizing each power generation component in the photovoltaic panel by the power generation efficiency optimization mechanism, the optimization adjustment feedback information of each power generation component is obtained in real time and preprocessed after acquisition;

[0121] During the optimization of each power generation component in the photovoltaic panel by the power generation efficiency optimization mechanism, real-time acquisition of optimization adjustment feedback information can be achieved through a combination of sensor data acquisition and optimization command response tracking. Specifically, each power generation component is equipped with a power output sensor, a current and voltage characteristic monitoring module, and a shading change monitoring module to record the actual power output, current and voltage characteristic changes, and shading status changes after the component's optimization adjustment in real time. This data is transmitted to the central control system via a wireless sensor network or data bus. Simultaneously, the adjustment commands in the optimization mechanism are also recorded in the software system as part of the feedback information and correlated with the collected real-time data for analysis. Furthermore, through a time synchronization module, the system can precisely correlate the optimized feedback data with the time point of executing the optimization command to capture the dynamic characteristics of the adjustment response, thereby generating a complete data stream of optimization adjustment feedback information.

[0122] The optimized output characteristic information and optimized response characteristic information are extracted from the optimized adjustment feedback information of each preprocessed power generation component, and analyzed after extraction to generate the optimized output consistency index and optimized response sensitivity coefficient of each power generation component.

[0123] Extracting optimized output and response characteristics from preprocessed optimization feedback information can be achieved through data classification, feature extraction, and dynamic analysis. Specifically, the feedback information is classified into power-related data (e.g., actual power output versus target power) and optimization adjustment-related data (e.g., adjustment magnitude versus response time). Key characteristics, such as output power difference, response time, and efficiency change rate, are extracted from each. Time series analysis is then used to model the dynamic trends of the data, further extracting optimized output and response characteristics. Finally, feature aggregation algorithms (e.g., weighted averaging) are used to integrate the data, generating complete optimization characteristic information to support subsequent calculations.

[0124] An optimization effect evaluation model is constructed based on the optimized output consistency index and optimized response sensitivity coefficient of each generated power generation component. The optimization evaluation coefficient of each power generation component is generated by weighted summation and analyzed after generation to evaluate whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component has reached the expectation. Based on the evaluation results, the power generation efficiency optimization mechanism is optimized.

[0125] In this embodiment, the logic for obtaining the optimized output consistency index and optimized response sensitivity coefficient of each power generation component is as follows:

[0126] The optimized output characteristic information is extracted from the preprocessed optimization adjustment feedback information of each power generation component. Specifically, this includes the actual output power of each power generation component at different times during a certain period of the optimization process, the magnitude of change in actual output power, and the target power value set for each power generation component by the power generation efficiency optimization mechanism. These are then calibrated as follows: , and , This indicates a period of time during the optimization process. Time of the first The actual output power of each power generation component This indicates a period of time during the optimization process. Time of the first The variation range of the actual output power of each power generation component The power generation efficiency optimization mechanism is represented as the first... The target power value set for each power generation component. , , and All are positive integers;

[0127] During the optimization process, real-time acquisition of the actual output power of each power generation component, the amplitude of actual output power variation, and the target power value set by the power generation efficiency optimization mechanism for each component can be achieved through sensor data acquisition and optimization system data integration. Specifically, firstly, high-precision power sensors installed at the output end of the power generation components acquire the actual output power data of each component in real time. The power sensors record the instantaneous output power value of the component at fixed time intervals and transmit the data to the central control system. Secondly, the system generates the amplitude of actual output power variation by performing differential calculations on the time series data of actual output power. This process is automatically processed by software, which can capture the dynamic characteristics of power fluctuations and update the amplitude of variation data in real time. In addition, the target power value in the power generation efficiency optimization mechanism is calculated and generated by the optimization model based on the component's operating status, environmental conditions, and historical operating data. The target power value is stored in the optimization system and is associated with the actual power data when optimization commands are issued, dynamically updating the target value setting. All these data acquisitions and calculations are seamlessly integrated through the software system. The system can automatically match and synchronize the data acquired by the sensors in real time with the target power value output by the optimization model, thereby forming complete optimization adjustment feedback information and providing accurate basic data support for subsequent analysis and optimization. This approach not only ensures the real-time nature and accuracy of data acquisition, but also effectively addresses the complex changes in component operating states under dynamic environments.

[0128] The optimal output consistency index of each power generation component is calculated using the following formula:

[0129]

[0130] In the formula, For the first The optimized output consistency index of each power generation component;

[0131] Calculate the optimal output consistency index of each power generation component. This is to comprehensively evaluate the consistency between the actual output and the target power, as well as the stability of the output power, during the optimization and adjustment process of the power generation components. Each calculation step in the formula aims to quantify the impact of different factors on the optimization effect. First, It is used to measure the relative deviation between actual output power and target power. The effect of larger deviations is amplified through squaring, thus highlighting moments when the target is not achieved satisfactorily. Secondly, This reflects the variation in actual output power over time, directly quantifying the volatility of the component output. Greater volatility indicates poorer stability of the optimized output. Accumulating these two components allows for simultaneous consideration of both output power accuracy and stability. Averaging over time periods normalizes the overall effect of the optimization feedback, ensuring comparability of optimization results for different components. Finally, taking the logarithm of the accumulated result smooths the impact of small deviations and fluctuations, while assigning higher weights to larger deviations and instabilities, highlighting significant issues in the optimization effect. This calculation logic comprehensively captures the consistency and stability of the component's optimized output, providing a scientific quantitative basis for subsequent optimization evaluation.

[0132] No. Optimized output consistency index of individual power generation components The magnitude of directly reflects the degree to which the power generation efficiency optimization mechanism achieves its optimization effect on the component, and is closely related to assessing whether the optimization effect has met expectations. When A smaller deviation indicates that the actual output power of the component deviates little from the target power, and the fluctuation range of the output power is low. This suggests that the optimized operation is relatively stable and the target is highly achieved, meaning the optimization effect basically meets expectations. Conversely, if... A large deviation indicates a significant discrepancy between the actual and target output power of the module, or noticeable fluctuations in output power. This suggests that the optimization mechanism has failed to effectively improve module performance, resulting in lower-than-expected optimization results. Therefore, by comparing... The size and pre-set optimization evaluation criteria can accurately assess whether the optimization effect has met expectations and provide a clear direction for further optimization.

[0133] The optimization response characteristic information is extracted from the preprocessed optimization adjustment feedback information of each power generation component. Specifically, this includes the percentage change in the ratio of actual input energy to output energy of each power generation component at different times during the optimization process, the rate of change in the power generation efficiency of each power generation component, and the time interval from receiving the optimization command to achieving the adjustment effect for each power generation component. These are then calibrated as follows: , and , This indicates a period of time during the optimization process. Time of the first The percentage change in the ratio of actual input energy to output energy of each power generation component. This indicates a period of time during the optimization process. Time of the first The rate of change in the power generation efficiency of each power generation component Indicates the first step in the optimization process. The time interval between receiving an optimization command and achieving the desired adjustment effect for each power generation component;

[0134] During the optimization process, real-time acquisition of the percentage change in the ratio of actual input energy to output energy, the rate of change in power generation efficiency, and the time interval from receiving the optimization command to achieving the adjustment effect for each power generation component can be achieved by combining a sensor network, a data acquisition module, and an optimization command tracking mechanism. Specifically, the percentage change in the ratio of actual input energy to output energy can be obtained collaboratively by current and voltage sensors and power sensors installed on the power generation components. The current and voltage sensors measure the input electrical energy (such as the product of the voltage and current received by the component), and the power sensors record the output power. The ratio is calculated by comparing the two through time series, and the rate of change is dynamically analyzed by software. The real-time acquisition of the rate of change in power generation efficiency is based on historical power output data and the power improvement after optimization. The software system calculates the efficiency difference at the same time before and after optimization and normalizes it into a rate of change to reflect the degree of improvement in power generation performance. As for the time interval from receiving the optimization command to achieving the adjustment effect, it can be accurately tracked through the command log and sensor feedback in the optimization mechanism. The software system records the sending time of each optimization command and the feedback time when the component's operating state stabilizes after optimization; the time difference is the response interval. All of these data acquisitions are integrated in real time through the data processing module in the central control system, and dynamic monitoring algorithms ensure efficient data acquisition and analysis, providing comprehensive and accurate basic information support for optimization and evaluation.

[0135] The optimized response sensitivity coefficient of each power generation component is calculated using the following formula:

[0136]

[0137] In the formula, For the first Optimized response sensitivity coefficients for each power generation component.

[0138] Calculate the optimal response sensitivity coefficient of each power generation component. This is to quantify the significance and timeliness of the component's response during optimization and adjustment. Each calculation step in the formula aims to comprehensively evaluate the multifaceted impact of optimization and adjustment on component performance. First, This ratio reflects the rate of change in power generation efficiency to the response delay time, and is used to measure the optimized efficiency of the module. The shorter the response time and the more significant the efficiency improvement, the larger the ratio, indicating a more ideal optimization effect; secondly, By taking the logarithm of the percentage change in the input-output energy ratio, the impact of larger changes is amplified while the contribution of smaller changes is smoothed, capturing the nonlinear effect of optimization adjustments on energy utilization. Multiplying these two parts, combined with the dynamic characteristics of optimization adjustments, comprehensively reflects the quality and speed of the optimization response. Finally, by averaging the data over all time points within the time period, the dynamic changes in the optimization response are unified into a comparable quantitative indicator. This calculation logic comprehensively integrates optimization efficiency, response timeliness, and energy utilization characteristics, providing a scientific and comprehensive basis for evaluating optimization effectiveness.

[0139] No. Optimized response sensitivity coefficient of each power generation component The magnitude of this value directly reflects whether the power generation efficiency optimization mechanism has achieved the expected optimization effect on this component. If... A larger value indicates that the component exhibits a higher rate of change in power generation efficiency after optimization, a shorter response time to optimization commands, and a significant improvement in the input-to-output energy ratio. Overall, this demonstrates that the optimization has a significant effect on improving component performance, and the optimization objectives have been largely achieved. Conversely, if... A smaller value indicates that the component's power generation efficiency improvement after optimization is limited, the response time is long, or the improvement in energy utilization is not significant, showing that the optimization effect is lower than expected. Therefore, The magnitude of the value is an important quantitative indicator for evaluating the effectiveness of the optimization mechanism. It can clearly reveal whether the optimization has effectively achieved the expected goal and provide a basis for subsequent adjustment of the optimization strategy.

[0140] In this embodiment, the optimized output consistency index of each generated power generation component is calculated. and optimize response sensitivity coefficient An optimization performance evaluation model is constructed, and optimization evaluation coefficients for each power generation component are generated through weighted summation. According to the formula: ,in and The optimized output consistency index for each power generation component and optimize response sensitivity coefficient The non-zero weight coefficients, and ;

[0141] Constructing an optimization performance evaluation model can be achieved by combining the optimization output consistency index. and optimize response sensitivity coefficient The weighted summation is achieved by using characteristic data extracted from the optimization and adjustment feedback information to calculate the weighted summation for each power generation component. and Then, based on the objective requirements of the optimization mechanism, weighting coefficients are set for these two indices. and The weighting coefficients are set based on the focus of the optimization objective: if more emphasis is placed on the consistency of the optimized output and the achievement of the objective, then weighting is assigned... Higher weighting; conversely, if more emphasis is placed on optimizing the responsiveness and efficiency of adjustments, then increase... The weights. The weight coefficients also need to satisfy... This is to ensure that the results of the evaluation model have physical meaning within a certain range. This is achieved through calculation formulas. The generated optimization evaluation coefficients It can comprehensively reflect the output consistency of components and also demonstrate their response efficiency, providing a scientific basis for the quantitative evaluation of optimization effects.

[0142] The optimized evaluation coefficients of each generated power generation component Compared with the pre-set optimization evaluation coefficient thresholds for each power generation component A comparison was conducted, and the optimization effect of the power generation efficiency optimization mechanism on each power generation component was evaluated based on the comparison results. The power generation efficiency optimization mechanism was then optimized based on the evaluation results. The specific comparative analysis is as follows:

[0143] like The power generation efficiency optimization mechanism has achieved the expected optimization effect on the power generation component, and there is no need to optimize the power generation efficiency optimization mechanism.

[0144] This indicates that the power generation efficiency optimization mechanism has achieved its expected optimization goals for the component. This means that after optimization, the component's output consistency and response sensitivity meet the performance requirements of the optimization mechanism; that is, the component's actual output power is close to the target power, with minimal power fluctuations, and its response to optimization commands is fast and efficient. The impact is that the component is already in an optimal state, requiring no further optimization, thus saving the optimization mechanism's computational resources and time, and ensuring the overall efficiency and stability of the photovoltaic equipment's operation.

[0145] like The power generation efficiency optimization mechanism has not achieved the expected optimization effect on the power generation component, and the mechanism needs to be optimized. Specifically, this includes: readjusting the operating parameter settings in the optimization strategy, such as current and voltage characteristic values ​​or local shading compensation parameters; improving the feedback information processing mechanism to enhance the real-time performance and accuracy of optimization adjustments; dynamically updating the optimization effect evaluation model to adapt it to current conditions based on new operating data; and optimizing the execution process to ensure that the power generation efficiency optimization mechanism can improve the power generation performance of each power generation component by reducing response latency and improving the efficiency of optimization command execution.

[0146] This situation indicates that the power generation efficiency optimization mechanism has failed to effectively improve the performance of the component, and the optimization effect has not met expectations. This suggests that the actual output power of the component deviates significantly from the target power, with noticeable power fluctuations, or that the response to optimization commands is slow, resulting in insufficient improvement in its power generation performance. The impact of this situation is that the component may operate in a suboptimal state for an extended period, affecting not only its own power generation efficiency but also potentially reducing the overall operating efficiency of the photovoltaic system. Therefore, targeted adjustments and improvements to the optimization mechanism are needed to ensure that the component can return to its optimal operating state while maintaining continuous improvement in the overall system performance.

[0147] When the optimization effect of the power generation efficiency optimization mechanism on a certain power generation component fails to meet expectations, the mechanism needs to be optimized through software. This includes the following aspects: First, the operating parameters are reset through the optimization strategy adjustment module. For example, the adjustment of current and voltage characteristic values ​​can be achieved by real-time monitoring of the component's maximum power point (MPP) position and dynamically recalculating the optimal operating point based on shading compensation data, ensuring the component operates at high efficiency under the current environment. Second, the feedback information processing mechanism is improved. This can be done by introducing data filtering algorithms (such as Kalman filtering) to enhance the accuracy of feedback data and by calibrating the real-time performance of optimization adjustments through timestamps, ensuring a more accurate dynamic correlation between feedback data and optimization instructions. Third, the dynamic update module of the optimization effect evaluation model is used to adjust model parameters online based on the latest operating data. This includes recalibrating the threshold range of optimization evaluation coefficients or adjusting weight parameters, enabling the model to adapt to changes in the component's operating environment. Finally, instruction queue optimization and priority management mechanisms are introduced into the optimization execution process to reduce execution latency. For example, parallel processing can improve instruction execution efficiency while prioritizing optimization tasks for high-impact components. These improvements, through automated processing and dynamic adjustment of the software, enable the optimization mechanism to adapt more flexibly to complex and ever-changing operating environments, not only improving optimization efficiency but also ensuring the stability and reliability of the overall performance of photovoltaic equipment.

[0148] The pre-set optimization evaluation coefficient thresholds for each power generation component can be determined by software based on a comprehensive analysis of historical operating data, environmental conditions, and performance targets. Specifically, this involves collecting historical optimization data of photovoltaic equipment under different operating environments (such as time-series information on the optimization output consistency index and optimization response sensitivity coefficient), and grouping the optimization effects of different components using cluster analysis (such as K-means clustering) to identify the characteristics of low-impact, medium-impact, and high-impact components. Combining environmental variables (such as light intensity, temperature, and shading mode) and target performance, a multivariate regression model is used to fit the optimization effects of each group, generating a reference range for optimization evaluation coefficients. Finally, the thresholds are dynamically adjusted using an empirical weighting method to ensure that they meet performance targets while adapting to different operating conditions, achieving a scientific and accurate setting of evaluation coefficients.

[0149] The impact of dynamic shading on photovoltaic panels, the effectiveness of optimization measures, and the overall operation data of photovoltaic equipment are comprehensively monitored and analyzed to continuously improve the power generation efficiency optimization mechanism and enhance the operating efficiency and dynamic adaptability of photovoltaic equipment.

[0150] To achieve comprehensive monitoring and analysis of the impact of dynamic shading on photovoltaic (PV) panels, the effectiveness of optimization measures, and overall PV equipment operation data, and to continuously improve the power generation efficiency optimization mechanism, this can be achieved by combining multi-source data acquisition, dynamic modeling, and a feedback optimization system. Specifically, firstly, a sensor network is used to collect key operational data of the PV equipment in real time, including the dynamic shading characteristics of the PV panels (such as the size and rate of change of the shading area), the output power and power fluctuations of each power generation component, and environmental parameters (such as light intensity and temperature), constructing a multi-dimensional data input system. Secondly, the dynamic modeling module in the software analyzes the collected data, for example, by using time series analysis and machine learning algorithms to model the impact of dynamic shading on power generation efficiency, and correlates this with the effectiveness of optimization measures to evaluate the applicability and improvement potential of the current optimization mechanism. Then, based on these analysis results, the system automatically generates improvement suggestions, such as adjusting optimization strategy parameters, optimizing response timing, or recalibrating model weights, dynamically updating the power generation efficiency optimization mechanism. Finally, combined with a global monitoring system, the analysis results and improvement measures of the overall operating status are fed back to the operation and maintenance team in real time through data visualization, enabling human-machine collaborative decision-making. This approach not only continuously improves the operating efficiency of photovoltaic equipment, but also enhances the system's dynamic adaptability to complex environmental changes, ensuring the long-term efficient operation and stability of photovoltaic equipment.

[0151] like Figure 2 The computer vision-based automated operation and maintenance management system for photovoltaic equipment shown includes a dynamic shading detection module, an operation status assessment module, a power generation efficiency optimization module, an optimization effect feedback module, and a comprehensive monitoring and improvement module.

[0152] The dynamic shading detection module uses computer vision technology to detect in real time whether dynamic shading occurs on the surface of the photovoltaic panel, which is the core power generation unit, during the operation of the photovoltaic equipment. When dynamic shading is detected on the surface of the photovoltaic panel, the module identifies the various power generation components in the photovoltaic panel that are used for power generation.

[0153] The operation status assessment module acquires the operation status information of each power generation component in the photovoltaic panel in real time when the photovoltaic panel is affected by dynamic shading, and analyzes the information to assess the degree of power generation efficiency reduction caused by dynamic shading of each power generation component in the photovoltaic panel, and classifies each power generation component in the photovoltaic panel into low-impact components, medium-impact components and high-impact components.

[0154] The power generation efficiency optimization module constructs a power generation efficiency optimization mechanism based on the division of each power generation component in the photovoltaic panel, and implements different optimization measures for low-impact, medium-impact, and high-impact components.

[0155] The optimization effect feedback module acquires the optimization adjustment feedback information of each power generation component in the photovoltaic panel in real time during the process of the power generation efficiency optimization mechanism optimizing each power generation component. After acquisition, it analyzes and evaluates whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component has met expectations, and optimizes the power generation efficiency optimization mechanism based on the evaluation results.

[0156] The comprehensive monitoring and improvement module comprehensively monitors and analyzes the impact of dynamic shading on photovoltaic panels, the effectiveness of optimization measures, and the overall operation data of photovoltaic equipment. It continuously improves the power generation efficiency optimization mechanism and enhances the operating efficiency and dynamic adaptability of photovoltaic equipment.

[0157] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0158] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. 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 data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0159] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0160] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0161] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0162] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0163] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0164] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A computer vision-based photovoltaic device automated operation and maintenance management method, characterized in that, Specifically comprising the following steps: During the operation of the photovoltaic device, the computer vision technology is used to detect in real time whether the photovoltaic panel surface as the core power generation unit appears a dynamic shielding phenomenon, and when the photovoltaic panel surface appears the dynamic shielding phenomenon, the power generation components for power generation in the photovoltaic panel are determined; The running state information of the power generation components in the photovoltaic panel during the process of being affected by the dynamic shielding is acquired in real time, and after being acquired, the running state information is analyzed, the power generation efficiency reduction degree of the power generation components in the photovoltaic panel caused by the dynamic shielding is evaluated, and the power generation components in the photovoltaic panel are divided into low-impact components, medium-impact components and high-impact components; According to the division results of the power generation components in the photovoltaic panel, a power generation efficiency optimization mechanism is constructed, and different optimization measures are taken for the low-impact components, the medium-impact components and the high-impact components; During the optimization of the power generation components in the photovoltaic panel by the power generation efficiency optimization mechanism, the optimization adjustment feedback information of the power generation components is acquired in real time, and after being acquired, the optimization adjustment feedback information is analyzed, whether the optimization effect of the power generation efficiency optimization mechanism on the power generation components reaches the expectation is evaluated, and the power generation efficiency optimization mechanism is optimized according to the evaluation results; The influence of the photovoltaic panel dynamic shielding, the implementation effect of the optimization measures and the overall operation data of the photovoltaic device are comprehensively monitored and analyzed, the power generation efficiency optimization mechanism is continuously improved, and the operation efficiency and the dynamic adaptation capability of the photovoltaic device are improved; The running state information of the power generation components in the photovoltaic panel during the process of being affected by the dynamic shielding is acquired in real time, and after being acquired, the running state information is analyzed, the power generation efficiency reduction degree of the power generation components in the photovoltaic panel caused by the dynamic shielding is evaluated, and the power generation components in the photovoltaic panel are divided into low-impact components, medium-impact components and high-impact components, specifically comprising the following steps: The running state information of the power generation components in the photovoltaic panel during the process of being affected by the dynamic shielding is acquired in real time, and after being acquired, the running state information is analyzed, the power generation efficiency reduction degree of the power generation components in the photovoltaic panel caused by the dynamic shielding is evaluated, and the power generation components in the photovoltaic panel are divided into low-impact components, medium-impact components and high-impact components, specifically comprising the following steps: The light shielding characteristic information and the power generation power fluctuation characteristic information in the preprocessed running state information are extracted, and after being extracted, the light shielding characteristic information and the power generation power fluctuation characteristic information are analyzed, and the shielding intensity index and the power generation power fluctuation coefficient of each power generation component are generated respectively; The shielding intensity index and the power generation power fluctuation coefficient of each power generation component are used to construct a power generation efficiency reduction evaluation model, the power generation evaluation coefficient of each power generation component is generated by weighted summation, and after being generated, the power generation evaluation coefficient is analyzed, the power generation efficiency reduction degree of the power generation components in the photovoltaic panel caused by the dynamic shielding is evaluated, and the power generation components in the photovoltaic panel are divided into low-impact components, medium-impact components and high-impact components; The acquisition logic of the shielding intensity index and the power generation power fluctuation coefficient of each power generation component is as follows: extracting the light-shading characteristic information in the preprocessed operating state information, specifically including the proportion of the shaded area to the total surface of each power generation component at different time points in a period of time during which the photovoltaic panel is affected by dynamic shading, the light intensity received by each power generation component, and the change rate of the light intensity received by each power generation component, and expressing the proportion, the light intensity, and the change rate respectively by functions 、 and in time sequence, is a time point, represents the proportion of the shaded area to the total surface of the i-th power generation component at time point t, represents the light intensity received by the i-th power generation component at time point t, represents the change rate of the light intensity received by the i-th power generation component at time point t, the definition period of time is , , , , is a positive integer;​​​ comparing the light intensity received by each power generation component at different time instants in a period of time during which the photovoltaic panel is affected by dynamic shading, and designating as the maximum value among them The shielding intensity index of each power generation component is calculated, and the specific calculation formula is as follows: ； In the formula, the shielding strength index of the nth power generation assembly; Extract the power generation fluctuation characteristics from the preprocessed operating status information. Specifically, this includes the actual power output value, the amplitude of change of the actual power output value, and the maximum power output value of each power generation module at different times during a period of time when the photovoltaic panel is affected by dynamic shading. Then, apply the actual power output value and the amplitude of change of the actual power output value of each power generation module at different times during a period of time when the photovoltaic panel is affected by dynamic shading to a function according to the time series. and To express, This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The actual power output value of each power generation component This indicates the period of time during which photovoltaic panels are affected by dynamic shading. Time of the first The variation in the actual power output value of each power generation component will be the first The maximum power output value of each power generation module is calibrated under standard test conditions as follows: ; The power generation power fluctuation coefficient of each power generation component is calculated, and the specific calculation formula is as follows: ； In the formula, For the first The power generation fluctuation coefficient of each power generation component. 2.The computer vision-based photovoltaic device automatic operation and maintenance management method according to claim 1, characterized in that, the shading intensity index of each power generation component and the power generation power fluctuation coefficient An evaluation model of power generation efficiency reduction is constructed, and the power generation evaluation coefficient of each power generation component is generated by weighted summation , according to the formula: , wherein and are the non-zero weight coefficients of the shading intensity index and the power generation power fluctuation coefficient of each power generation component, and ; The power generation evaluation coefficients of each generated power generation component Compared with the pre-set power generation assessment coefficient threshold range A comparison was conducted, and the degree of power generation efficiency reduction caused by dynamic shading in each photovoltaic module was assessed based on the comparison results. The photovoltaic modules were then categorized into low-impact, medium-impact, and high-impact modules. The specific comparison analysis and categorization are as follows: If , the power generation efficiency of the power generation component in the photovoltaic panel caused by dynamic shading is low, and the power generation component is divided into a low-impact component; If , the power generation efficiency of the power generation component in the photovoltaic panel caused by dynamic shading is moderate, and the power generation component is divided into a moderate-impact component; If , the photovoltaic panel is divided into high-impact components. 3.The computer vision-based photovoltaic device automatic operation and maintenance management method of claim 2, wherein, According to the division result of each power generation component in the photovoltaic panel, a power generation efficiency optimization mechanism is constructed, specifically: according to the division result of low-impact components, medium-impact components and high-impact components, different optimization adjustment parameters are set to form a power generation efficiency optimization mechanism; the optimization mechanism is based on the power generation evaluation coefficient of each power generation component and the dynamic feedback of the shading intensity index and the power generation power fluctuation coefficient, and through the pre-set optimization rule, the operation parameter adjustment mode and adjustment range of the component are automatically determined; Different optimization measures are taken for low-impact components, medium-impact components and high-impact components, specifically: The optimization measure for low-impact components is to use the operation maintenance parameters in the optimization mechanism to maintain the current operation state without parameter adjustment; The optimization measure for medium-impact components is to use the operation adjustment parameters in the optimization mechanism to adjust the operation parameters of the power generation component, including resetting the current-voltage characteristic value and implementing the shading compensation strategy, so as to reduce the influence of dynamic shading on the power generation performance; The optimization measure for high-impact components is to use the operation optimization parameters in the optimization mechanism to adjust the operation mode of the power generation component, including dynamically changing the distribution of shading areas, redistributing the component operation load and implementing full coverage compensation measures for the shading area, so as to ensure the recovery of power generation performance. 4.The computer vision-based photovoltaic device automatic operation and maintenance management method of claim 3, wherein, In the process of optimizing each power generation component in the photovoltaic panel by the power generation efficiency optimization mechanism, the optimization adjustment feedback information of each power generation component is obtained in real time, and after being obtained, it is analyzed to evaluate whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component reaches the expectation, and the power generation efficiency optimization mechanism is optimized according to the evaluation result, specifically including the following steps: In the process of optimizing each power generation component in the photovoltaic panel by the power generation efficiency optimization mechanism, the optimization adjustment feedback information of each power generation component is obtained in real time, and after being obtained, it is analyzed to evaluate whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component reaches the expectation, and the power generation efficiency optimization mechanism is optimized according to the evaluation result, specifically including the following steps: The optimization output characteristic information and optimization response characteristic information in the optimization adjustment feedback information of each power generation component after preprocessing are extracted, and after being extracted, they are analyzed to generate the optimization output consistency index and optimization response sensitivity coefficient of each power generation component; An optimization effect evaluation model is constructed for the generated optimization output consistency index and optimization response sensitivity coefficient of each power generation component, a weighted sum is generated to generate the optimization evaluation coefficient of each power generation component, and after being generated, it is analyzed to evaluate whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component reaches the expectation, and the power generation efficiency optimization mechanism is optimized according to the evaluation result. 5.The computer vision-based photovoltaic device automatic operation and maintenance management method according to claim 4, characterized in that, The acquisition logic of the optimization output consistency index and optimization response sensitivity coefficient of each power generation component is as follows: The optimization output characteristic information in the optimization adjustment feedback information of each power generation component after preprocessing is extracted, and specifically includes the actual output power of each power generation component at different time points within a period of time in the optimization process, the change amplitude of the actual output power, and the target power value set for each power generation component by the power generation efficiency optimization mechanism, and is respectively labeled as , and , represents the actual output power of the i-th power generation component at the t-th time point within a period of time in the optimization process, represents the change amplitude of the actual output power of the i-th power generation component at the t-th time point within a period of time in the optimization process, represents the target power value set for the i-th power generation component by the power generation efficiency optimization mechanism, , , , and are positive integers;​​​​ The optimization output consistency index of each power generation component is calculated, and the specific calculation formula is as follows: ； In the formula, is the optimization output consistency index of the nth power generation assembly; The optimization response characteristic information is extracted from the preprocessed optimization adjustment feedback information of each power generation component. Specifically, this includes the percentage change in the ratio of actual input energy to output energy of each power generation component at different times during the optimization process, the rate of change in the power generation efficiency of each power generation component, and the time interval from receiving the optimization command to achieving the adjustment effect for each power generation component. These are then calibrated as follows: , and , This indicates a period of time during the optimization process. Time of the first The percentage change in the ratio of actual input energy to output energy of each power generation component. This indicates a period of time during the optimization process. Time of the first The rate of change in the power generation efficiency of each power generation component Indicates the first step in the optimization process. The time interval between receiving an optimization command and achieving the desired adjustment effect for each power generation component; The optimization response sensitivity coefficient of each power generation component is calculated, and the specific calculation formula is as follows: ； In the formula, is the optimized response sensitivity coefficient of the nth power generation assembly. 6.The computer vision-based photovoltaic device automatic operation and maintenance management method according to claim 5, characterized in that, Optimized output consistency index of each power generation component And optimized response sensitivity coefficient An optimized effect evaluation model is constructed, and an optimized evaluation coefficient of each power generation component is generated by weighted summation , according to the formula: , wherein And The non-zero weight coefficient of the optimized output consistency index And optimized response sensitivity coefficient of each power generation component, and ; The generated optimization evaluation coefficient of each power generation component The preset optimization evaluation coefficient threshold of each power generation component The comparison is made, and according to the comparison result, it is evaluated whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component reaches the expectation, and the power generation efficiency optimization mechanism is optimized according to the evaluation result. The specific comparison analysis is as follows: If , the optimization effect of the power generation efficiency optimization mechanism on the power generation assembly reaches the expectation, and the power generation efficiency optimization mechanism does not need to be optimized. If The power generation efficiency optimization mechanism needs to be optimized if the optimization effect of the power generation assembly does not meet the expected effect, including: re-adjusting the operating parameter settings in the optimization strategy; improving the feedback information processing mechanism to improve the real-time and accuracy of optimization adjustment; dynamically updating the optimization effect evaluation model to adapt to the current conditions based on new operation data; optimizing the execution process to ensure that the power generation efficiency optimization mechanism can improve the power generation performance of each power generation assembly by reducing response delay and improving optimization instruction execution efficiency.

7. A computer vision-based photovoltaic device automated operation and maintenance management system for implementing the computer vision-based photovoltaic device automated operation and maintenance management method of any one of claims 1-6, characterized in that, It includes a dynamic shading detection module, an operation state evaluation module, a power generation efficiency optimization module, an optimization effect feedback module and a comprehensive monitoring improvement module; The dynamic occlusion detection module uses computer vision technology to detect whether the photovoltaic panel surface as the core power generation unit appears dynamic occlusion phenomenon in real time during the operation of the photovoltaic device, and determines each power generation component for power generation in the photovoltaic panel when the dynamic occlusion phenomenon appears on the photovoltaic panel surface; The running state evaluation module obtains the running state information of each power generation component in the photovoltaic panel in the process of being affected by the dynamic occlusion of the photovoltaic panel in real time, analyzes after obtaining, evaluates the power generation efficiency reduction degree of each power generation component in the photovoltaic panel caused by the dynamic occlusion, and divides each power generation component in the photovoltaic panel into low-impact components, medium-impact components and high-impact components; The power generation efficiency optimization module constructs a power generation efficiency optimization mechanism according to the division results of each power generation component in the photovoltaic panel, and takes different optimization measures for low-impact components, medium-impact components and high-impact components; The optimization effect feedback module obtains the optimization adjustment feedback information of each power generation component in real time in the process of the power generation efficiency optimization mechanism optimizing each power generation component in the photovoltaic panel, analyzes after obtaining, evaluates whether the optimization effect of the power generation efficiency optimization mechanism on each power generation component reaches the expectation, and optimizes the power generation efficiency optimization mechanism according to the evaluation result; The comprehensive monitoring and improvement module comprehensively monitors and analyzes the influence of the photovoltaic panel dynamic occlusion, the implementation effect of the optimization measure and the overall operation data of the photovoltaic device, continuously improves the power generation efficiency optimization mechanism, and improves the operation efficiency and dynamic adaptability of the photovoltaic device.

Images