Power system anomaly detection method and system

Abstract

The invention provides a power system anomaly detection method and system, and the method comprises the steps: carrying out the prediction of a photovoltaic panel assembly based on real-time environment parameters collected in real time and a constructed power prediction model, and obtaining theoretical power generation power data; comparing the obtained real-time power generation power data of the photovoltaic panel assembly with the theoretical power generation power data to obtain a comparison result, and determining an abnormal photovoltaic panel assembly; and performing analysis based on the image data of the abnormal photovoltaic panel assembly acquired in real time to obtain an abnormal result. According to the method, by comparing the real-time power generation power and the theoretical power generation power of the photovoltaic panel assembly, the photovoltaic panel assembly with the abnormal power generation efficiency can be rapidly identified, potential problems can be directly, effectively and timely found, energy waste and potential safety hazards caused by reduction of the power generation efficiency are avoided, and the power generation efficiency of the photovoltaic panel assembly is improved. By acquiring and analyzing the image data of the photovoltaic panel assembly in real time, the actual condition of the photovoltaic panel can be known more visually, and the accuracy of anomaly detection is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power detection, and in particular to a method and system for detecting abnormality in a power system. Background Art

[0002] With the increasing use of renewable energy and the rapid advancement of photovoltaic technology around the world, photovoltaic power stations have gradually become a key link in the transformation of the global energy structure. These power stations provide important support for the sustainable development of society by efficiently converting solar energy into electrical energy. However, the operating environment of photovoltaic power stations is extremely complex and changeable. They not only have to face unpredictable weather conditions, such as high temperature, severe cold, strong winds, heavy rain and other extreme weather conditions, but also have to deal with dust accumulation, bird droppings pollution and possible human damage and other potential threats. These external factors often cause damage or failure to key equipment of photovoltaic power stations such as photovoltaic panel components, inverters, and box transformers, thereby affecting the normal operation of the entire power system and the stability of power supply.

[0003] At present, in order to cope with the extremely complex and changeable environment in which photovoltaic power stations are located, detection methods based on mechanism models, data-driven detection methods and image recognition-based detection methods have been proposed; among them, the detection method based on mechanism models attempts to predict and detect abnormal situations by establishing an accurate mathematical model of the power system; but the complexity and uncertainty of photovoltaic power stations make the establishment of such models extremely difficult. At the same time, the operation of photovoltaic power stations involves multiple physical processes and influencing factors. The interactions and dependencies between these factors are often difficult to accurately describe with simple mathematical models; and the detection method based on data-driven relies on a large amount of historical data for pattern recognition and anomaly detection. However, the data collection and storage costs of photovoltaic power stations are usually high, and the accuracy and integrity of the data are difficult to be fully guaranteed. In addition, due to the changeable operating environment of photovoltaic power stations, historical data may It cannot fully reflect the current or future actual situation, thus affecting the accuracy and reliability of the detection method; and the detection method based on image recognition detects abnormal situations by capturing and analyzing the image information of photovoltaic power station equipment. Although it can intuitively reflect the operating status of the equipment and changes in the external environment, image recognition technology is also easily affected by various factors such as lighting conditions, shooting angles, and obstructions in actual applications. These factors often lead to a decrease in image quality and a decrease in recognition accuracy, thus affecting the stability and reliability of the detection method; therefore, the existing methods for detecting abnormalities in power systems in photovoltaic power stations are difficult to fully adapt to the variability and uncertainty of the operating environment of photovoltaic power stations, resulting in a decrease in the accuracy and reliability of prediction and detection results, and cannot accurately and reliably detect abnormal situations in photovoltaic power stations. In severe cases, it is easy to paralyze the power system in the photovoltaic power station. Summary of the invention

[0004] In order to solve the problem that the existing power system anomaly detection method is difficult to fully adapt to the variability and uncertainty of the operating environment of photovoltaic power stations, resulting in reduced accuracy and reliability of prediction and detection results, the present invention provides a power system anomaly detection method and system.

[0005] To achieve the above object, the present invention provides the following technical solutions: The present invention proposes a method for detecting abnormality in a power system, comprising the following steps: Based on the real-time environmental parameters collected in real time and the constructed power prediction model, the photovoltaic panel components are predicted to obtain theoretical power generation data; Based on the comparison between the real-time power generation data of the photovoltaic panel components and the theoretical power generation data, the comparison results are obtained to determine the abnormal photovoltaic panel components; Based on the image data of abnormal photovoltaic panel components acquired in real time, analysis is performed to obtain abnormal results.

[0006] Preferably, the prediction of photovoltaic panel components based on the real-time environmental parameters collected in real time and the constructed power prediction model includes: Build a power prediction model based on historical data of photovoltaic power plants; Collecting environmental parameters around the photovoltaic panel assembly in real time to obtain real-time environmental parameters; The real-time environmental data is input into a power prediction model for prediction to obtain theoretical power generation data.

[0007] Preferably, the building of the power prediction model includes: Acquire historical power generation data of the photovoltaic panel assembly and historical environmental parameters corresponding to the historical power generation to obtain historical data; A preliminary power prediction model is constructed based on the historical data and real-time environmental parameters; Extract a portion of the historical data to form a training data set and a verification data set; Inputting the training data set into the preliminary power prediction model to learn the mapping relationship between environmental parameters and generated power; Adjust the preliminary power prediction model based on the mapping parameters to obtain a power prediction optimization model; Inputting the verification data set into the power prediction optimization model for verification to obtain verification evaluation parameters; The power prediction optimization model is adjusted based on the verification evaluation parameters to obtain the power prediction model.

[0008] Preferably, the environmental parameters include sunshine intensity parameters, temperature parameters, temperature parameters and air flow parameters; the air flow parameters include wind speed and wind direction.

[0009] Preferably, the real-time power generation data of the photovoltaic panel assembly is obtained and compared with the theoretical power generation data to obtain a comparison result, including: Obtain the real-time power generation of photovoltaic panels; Calculate the difference between the real-time power generation data and the theoretical power generation data to obtain a comparison difference; The comparison difference is compared with a preset threshold range to obtain a comparison result.

[0010] Preferably, the analyzing the abnormal photovoltaic panel assembly based on the image data of the abnormal photovoltaic panel assembly acquired in real time includes: Collecting image data of all abnormal photovoltaic panel components; Inputting the image data into the constructed occlusion model for matching analysis to obtain a first detection result; An abnormality analysis is performed based on the image data and the first detection result to obtain an abnormality detection result of an abnormally operating photovoltaic panel assembly.

[0011] Preferably, the occluder model includes: Obtain an obstruction image of a historical obstruction on a photovoltaic panel assembly, manually annotate the obstruction image, and perform feature extraction on the manually annotated obstruction image to obtain obstruction feature data; The occluder image model is constructed based on the occluder feature data.

[0012] Preferably, performing abnormality analysis based on the image data and the first detection result includes: Collecting image data of all abnormal photovoltaic panel components; Inputting the image data into the constructed occlusion model for matching analysis to obtain a first detection result; An abnormality analysis is performed based on the image data and the first detection result to obtain an abnormality detection result of an abnormally operating photovoltaic panel assembly.

[0013] Preferably, performing abnormality analysis based on the image data and the first detection result includes: Performing grayscale processing on the image data to obtain a grayscale image; Performing peak information statistics on the grayscale image to obtain peak information; Performing threshold segmentation on the grayscale image after corrosion and expansion based on the peak number information to obtain a binary image; Performing corrosion and expansion on the binary image to filter out redundant boundary noise points, performing dot multiplication of the filtered binary image and the image after image delay processing of the finite impulse response filtered grayscale image to obtain a preprocessed image; The abnormal result is analyzed based on the abnormal area in the preprocessed image and the first detection result to obtain the abnormal result.

[0014] The present invention proposes a power system anomaly detection method for implementing the above detection method, including an acquisition module, an output module, a prediction module, a comparison module and an analysis module; The acquisition module is used to acquire the real-time environmental parameters and the real-time power generation data of the photovoltaic panel assembly collected in real time; The prediction module is used to predict the photovoltaic panel components based on the real-time environmental parameters collected in real time and the constructed power prediction model to obtain theoretical power generation data; The comparison module is used to obtain the real-time power generation data of the photovoltaic panel assembly and compare it with the theoretical power generation data to obtain the comparison result and determine the abnormal photovoltaic panel assembly; The analysis module is used to analyze the image data of abnormal photovoltaic panel components acquired in real time to obtain abnormal results; The output module is used to output abnormal results.

[0015] Compared with the prior art, the present invention has the following beneficial technical effects: The present invention proposes a method for detecting abnormalities in a power system. The method constructs a power prediction model and predicts the theoretical power generation of a photovoltaic panel assembly based on real-time environmental parameters, effectively utilizing currently available environmental information, thereby being able to more accurately reflect the power generation capacity of a photovoltaic panel under ideal conditions. This not only improves the accuracy of power generation prediction, but also can quickly identify photovoltaic panel assemblies with abnormal power generation efficiency by comparing the real-time power generation of the photovoltaic panel assembly with the theoretical power generation, thereby directly and effectively being able to promptly discover potential problems and avoid energy waste and safety hazards caused by reduced power generation efficiency. By acquiring and analyzing the image data of the photovoltaic panel assembly in real time, the actual condition of the photovoltaic panel can be more intuitively understood, thereby improving the accuracy of abnormality detection.

[0016] Furthermore, this method utilizes the rich historical data of photovoltaic power stations to construct a power prediction model, which not only considers the historical power generation data of photovoltaic panel components, but also integrates the corresponding historical environmental parameters, such as light intensity, temperature, humidity, etc., so as to more comprehensively reflect the power generation characteristics of photovoltaic panels in different environments. In the process of model construction, the division of training data set and verification data set is adopted. Through continuous iteration and optimization, the prediction accuracy of the model is gradually improved, ensuring that the power prediction model can still maintain a high prediction accuracy when facing complex and changeable actual environments. The environmental parameters around the photovoltaic panel components are collected in real time and input into the power prediction model for prediction, thereby obtaining theoretical power generation data, realizing real-time monitoring of the power generation efficiency of photovoltaic panels, and providing accurate benchmark values ​​for subsequent anomaly detection.

[0017] Furthermore, the method calculates the difference between the real-time power generation of the photovoltaic panel assembly and the theoretical power generation, and compares it with the preset threshold range, thereby realizing the accurate identification of abnormal power generation efficiency of the photovoltaic panel assembly, simplifying the abnormality detection process, and improving the accuracy and efficiency of detection. By acquiring the power generation data of the photovoltaic panel assembly in real time and calculating the difference with the theoretical predicted value, the deviation of the power generation efficiency can be quickly discovered. At the same time, the setting of the preset threshold range makes the abnormality detection more objective and scientific, and avoids the subjectivity and uncertainty of human judgment.

[0018] Furthermore, the obstruction model constructed by the present method can realize rapid matching analysis of obstructions on the current abnormal photovoltaic panel components based on the feature data of historical obstruction images, thereby effectively identifying the abnormal power generation efficiency caused by the obstructions and improving the pertinence of abnormality detection. The image data of the abnormal photovoltaic panel components are preprocessed and optimized by using grayscale processing, peak information statistics, threshold segmentation, corrosion expansion, etc., which effectively filters out redundant boundary noise points, improves image quality, and provides clearer and more accurate image information for subsequent abnormality analysis. By performing point multiplication operation on the preprocessed image and the finite impulse response filtered grayscale image after image delay processing, a more refined abnormal area image is obtained, which not only enhances the contrast of the abnormal area, but also provides strong support for the subsequent abnormal result analysis.

[0019] The present invention proposes a power system anomaly detection system. The system can collect real-time power generation data and real-time environmental parameters of photovoltaic panel components in real time through an acquisition module, providing a rich and accurate data basis for subsequent prediction and comparison, ensuring the real-time and accuracy of the detection system, so that abnormal problems can be discovered and processed in time. The prediction module uses the constructed power prediction model to predict the theoretical power generation of photovoltaic panel components based on real-time environmental parameters, taking into account the influence of environmental factors, and also using the accumulation of historical data to improve the accuracy of the prediction. By comparing with the real-time power generation data, the comparison module can quickly identify photovoltaic panel components with abnormal power generation efficiency. The analysis module introduces image processing technology to conduct in-depth analysis and accurate identification of image data of abnormal photovoltaic panel components, improves the intelligent level of anomaly detection, and provides a clear direction and basis for subsequent maintenance processing. The output module presents the abnormal results in an intuitive and clear manner, which is convenient for operation and maintenance personnel to understand and process in time, thus ensuring the practicability and operability of the detection system and improving the overall operation and maintenance efficiency of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 The present invention is a flowchart of a method for detecting abnormality in a power system. DETAILED DESCRIPTION

[0021] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and descriptions are considered to be exemplary and non-restrictive in nature.

[0022] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0023] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0024] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, an electrical connection, or a communication; it can be a direct connection, or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0025] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may include that the first and second features are in direct contact, or may include that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, a first feature being "above", "above" and "above" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is higher in level than the second feature. A first feature being "below", "below" and "below" a second feature includes that the first feature is directly above and obliquely above the second feature, or simply indicates that the first feature is lower in level than the second feature.

[0026] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0027] See also Figure 1 The present invention proposes a method for detecting abnormality in a power system, the method comprising: The present invention proposes a power system anomaly detection system, which is used to implement the above method, including: Based on the real-time environmental parameters collected in real time and the constructed power prediction model, the photovoltaic panel components are predicted to obtain theoretical power generation data; Specifically, a power prediction model is constructed based on the historical data of the photovoltaic power station; Obtain historical power generation data of photovoltaic panel components in a photovoltaic power station and historical environmental parameters corresponding to the historical power generation to obtain historical data; construct a preliminary power prediction model using a machine learning or deep learning algorithm for the historical data and real-time environmental parameters; extract a portion of data from the historical data to form a training data set and a verification data set; input the training data set into the preliminary power prediction model to learn the mapping relationship between environmental parameters and power generation to obtain mapping parameters, adjust the preliminary power prediction model through the mapping parameters to obtain a power prediction optimization model; input the verification data set into the power prediction optimization model for verification to obtain verification evaluation parameters, adjust the power prediction optimization model through the verification evaluation parameters to obtain a power prediction model; Real-time collection of environmental parameters around the photovoltaic panel assembly to obtain real-time environmental parameters; that is, the sunshine intensity of the photovoltaic panel assembly is collected through the photoelectric sensor or radiometer installed on the photovoltaic panel assembly to obtain the sunshine intensity parameter; the temperature of the photovoltaic panel surface or the surrounding environment is monitored through the temperature sensor installed on the photovoltaic panel assembly to obtain the temperature parameter; the humidity of the photovoltaic panel surface or the surrounding environment is monitored through the humidity sensor installed on the photovoltaic panel assembly to obtain the humidity parameter; the wind speed and wind direction of the photovoltaic panel surface or the surrounding environment is monitored through the anemometer installed on the photovoltaic panel assembly to obtain the air flow parameter; the sunshine intensity parameter, temperature parameter, humidity parameter and air flow parameter are integrated to obtain the real-time environmental parameters; The real-time environmental data is input into the power prediction model for prediction to obtain the theoretical power generation data.

[0028] Based on the comparison between the real-time power generation data of the photovoltaic panel components and the theoretical power generation data, the comparison results are obtained, and the abnormal photovoltaic panel components are preliminarily determined through the comparison results; Specifically, the real-time power generation of the photovoltaic panel assembly is obtained; the actual output power of the photovoltaic panel assembly under the corresponding real-time environmental data is measured by a power meter installed at the power output end of the photovoltaic panel assembly, so as to obtain the real-time power generation data of the photovoltaic panel assembly corresponding to the real-time environmental data; the real-time power generation data and the theoretical power generation data are difference calculated to obtain a comparison difference, the absolute value of the comparison difference is taken to obtain the absolute value of the difference, the absolute value of the difference is compared with a preset threshold range to obtain a comparison result, and the comparison result is: when the absolute value exceeds the preset threshold range, it is determined that the corresponding photovoltaic panel assembly is operating abnormally, and when the absolute value is within the preset threshold range, it is determined that the corresponding photovoltaic panel assembly is operating normally, and an abnormal photovoltaic panel assembly is obtained.

[0029] Among them, the preset threshold range is obtained by obtaining the historical output power and theoretical output power of the photovoltaic panel components in normal operation in the history, calculating the difference between the historical output power and the theoretical output power, obtaining a historical difference data set, taking the absolute value of all historical difference data in the difference data set, selecting the maximum historical difference data and the minimum historical difference data in the historical difference data after the absolute value, and constructing the preset threshold range with the maximum historical difference data as the upper limit and the minimum historical difference data as the lower limit.

[0030] The abnormal photovoltaic panel components are analyzed based on the image data of the abnormal photovoltaic panel components acquired in real time to obtain abnormal results.

[0031] Specifically, sorting out all abnormal photovoltaic panel assemblies, and using an image acquisition device to acquire image data of these photovoltaic panel assemblies; wherein the image data includes visible light images and infrared light images; The image data is input into the constructed obstruction model for matching analysis to detect whether there is an obstruction on the photovoltaic panel assembly, and obtain a first detection result; the first detection result includes the location and type of the obstruction.

[0032] The construction of the occlusion model includes: Obtain an obstruction image of historical obstructions on the photovoltaic panel assembly, manually annotate the obstruction image, annotate the type of obstruction, such as leaves, dust, bird droppings, etc., then perform feature extraction on the manually annotated obstruction image to obtain obstruction feature data, and construct an obstruction image model based on the obstruction feature data; An abnormality analysis is performed based on the image data and the first detection result to obtain an abnormal detection result of the abnormally operating photovoltaic panel assembly.

[0033] Specifically, grayscale processing is performed on the image data to obtain a grayscale image; The peak information of the grayscale image is counted to obtain the peak information; wherein the peak information includes the starting position, the ending position, the maximum grayscale value and the average grayscale value; specifically, the two widest groups of peak information in a column of grayscale images are counted, that is, the starting position and the stopping position of two consecutive groups of background pixel points in the grayscale image are counted, and the peak range is determined by the starting position and the stopping position. At the same time, the grayscale value of the background pixel at the peak position is calculated by the average grayscale value of all pixels less than a certain grayscale value in the previous frame of the image; Based on the peak number information, the grayscale image after corrosion and expansion is threshold segmented to obtain a binary image; Specifically, find the edge points that meet the conditions within the two groups of wave peaks in the eroded and expanded edge point image, and obtain the grayscale values ​​of the background pixels where these edge points are located; The grayscale value average is calculated through these grayscale values, and the calculated grayscale value average is used as the segmentation threshold, and two sets of segmentation thresholds are generated for each column of images; among which, the grayscale value of the background pixel is the average grayscale value of all pixels in the previous frame image that are less than the set grayscale value.

[0034] Based on two sets of peak ranges and two sets of segmentation thresholds, the edge image after corrosion and expansion is threshold segmented to obtain a binary image, that is, the pixels in the image are divided into two categories: foreground area (abnormal area) and background area (normal area); The binary image is corroded and expanded to filter out redundant boundary noise points, and the filtered binary image is dot-multiplied with the image after image delay processing of the finite impulse response filtered grayscale image to obtain a preprocessed image; analyzing the abnormal result based on the abnormal region in the preprocessed image and the first detection result; When the area ratio of the abnormal area in the preprocessed image matches the power ratio interval, the temperature difference between the highest temperature in the abnormal area and the lowest temperature in the working area is calculated; if the temperature difference is less than the preset temperature threshold, it is determined that some circuits in the photovoltaic panel assembly are disconnected and not working; if the temperature difference is greater than or equal to the temperature threshold, it is determined that a hot spot phenomenon occurs in the photovoltaic panel assembly.

[0035] When the area ratio of the abnormal region in the preprocessed image does not match the power ratio interval, and the output result of the obstruction detection model indicates that there is no obstruction on the surface of the photovoltaic panel, it is determined that there is dust scattered on the photovoltaic panel assembly.

[0036] The present invention also discloses a power system anomaly detection method for implementing the above method. The system includes an acquisition module, an output module, a prediction module, a comparison module and an analysis module. The acquisition module is communicatively connected to the prediction module, the prediction module is communicatively connected to the comparison module, the comparison module is communicatively connected to the analysis module, and the analysis module is communicatively connected to the output module. The acquisition module is used to obtain the real-time environmental parameters and real-time power generation data of the photovoltaic panel components collected in real time; The prediction module is used to predict the photovoltaic panel components based on the real-time environmental parameters collected in real time and the constructed power prediction model to obtain theoretical power generation data; The comparison module is used to compare the real-time power generation data of photovoltaic panels with the theoretical power generation data to obtain the comparison results and identify abnormal photovoltaic panels. The analysis module is used to analyze the image data of abnormal photovoltaic panel components acquired in real time to obtain abnormal results; The output module is used to output abnormal results.

[0037] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0038] In addition, it should be understood that although this specification is described in accordance with the implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation modes that can be understood by those skilled in the art. The above content is only to illustrate the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution according to the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A method for detecting abnormality in a power system, characterized in that: The following steps are involved: Based on the real-time environmental parameters collected in real time and the constructed power prediction model, the photovoltaic panel components are predicted to obtain theoretical power generation data; Based on the comparison between the real-time power generation data of the photovoltaic panel components and the theoretical power generation data, the comparison results are obtained to determine the abnormal photovoltaic panel components; Based on the image data of abnormal photovoltaic panel components acquired in real time, analysis is performed to obtain abnormal results.

2. A method for detecting abnormality in a power system according to claim 1, characterized in that: The prediction of photovoltaic panel components based on the real-time environmental parameters collected in real time and the constructed power prediction model includes: Build a power prediction model based on historical data of photovoltaic power plants; Collecting environmental parameters around the photovoltaic panel assembly in real time to obtain real-time environmental parameters; The real-time environmental data is input into a power prediction model for prediction to obtain theoretical power generation data.

3. A method for detecting abnormality in a power system according to claim 2, characterized in that: The constructing of the power prediction model comprises: Acquire historical power generation data of the photovoltaic panel assembly and historical environmental parameters corresponding to the historical power generation to obtain historical data; A preliminary power prediction model is constructed based on the historical data and real-time environmental parameters; Extract a portion of the historical data to form a training data set and a verification data set; Inputting the training data set into the preliminary power prediction model to learn the mapping relationship between environmental parameters and generated power; Adjust the preliminary power prediction model based on the mapping parameters to obtain a power prediction optimization model; Inputting the verification data set into the power prediction optimization model for verification to obtain verification evaluation parameters; The power prediction optimization model is adjusted based on the verification evaluation parameters to obtain the power prediction model.

4. A method for detecting abnormality in a power system according to claim 2, characterized in that: The environmental parameters include sunshine intensity parameters, temperature parameters, temperature parameters and air flow parameters; the air flow parameters include wind speed and wind direction.

5. A method for detecting anomalies in a power system according to claim 1, characterized in that: The real-time power generation data of the photovoltaic panel assembly is obtained and compared with the theoretical power generation data to obtain a comparison result, including: Obtain the real-time power generation of photovoltaic panels; Calculate the difference between the real-time power generation data and the theoretical power generation data to obtain a comparison difference; The comparison difference is compared with a preset threshold range to obtain a comparison result.

6. A method for detecting abnormality in a power system according to claim 1, characterized in that: The analyzing of abnormal photovoltaic panel components based on the image data of abnormal photovoltaic panel components acquired in real time includes: Collecting image data of all abnormal photovoltaic panel components; Inputting the image data into the constructed occlusion model for matching analysis to obtain a first detection result; An abnormality analysis is performed based on the image data and the first detection result to obtain an abnormality detection result of an abnormally operating photovoltaic panel assembly.

7. A method for detecting abnormality in a power system according to claim 6, characterized in that: The occlusion model includes: Obtain an obstruction image of a historical obstruction on a photovoltaic panel assembly, manually annotate the obstruction image, and perform feature extraction on the manually annotated obstruction image to obtain obstruction feature data; The occluder image model is constructed based on the occluder feature data.

8. A method for detecting abnormality in a power system according to claim 6, characterized in that: The performing abnormality analysis based on the image data and the first detection result includes: Collecting image data of all abnormal photovoltaic panel components; Inputting the image data into the constructed occlusion model for matching analysis to obtain a first detection result; An abnormality analysis is performed based on the image data and the first detection result to obtain an abnormality detection result of an abnormally operating photovoltaic panel assembly.

9. A method for detecting abnormality in a power system according to claim 6, characterized in that: The performing abnormality analysis based on the image data and the first detection result includes: Performing grayscale processing on the image data to obtain a grayscale image; Performing peak information statistics on the grayscale image to obtain peak information; Performing threshold segmentation on the grayscale image after corrosion and expansion based on the peak number information to obtain a binary image; Performing corrosion and expansion on the binary image to filter out redundant boundary noise points, performing dot multiplication of the filtered binary image and the image after image delay processing of the finite impulse response filtered grayscale image to obtain a preprocessed image; The abnormal result is analyzed based on the abnormal area in the preprocessed image and the first detection result to obtain the abnormal result.

10. A power system anomaly detection method for implementing the detection method according to any one of claims 1 to 9, characterized in that: It includes acquisition module, output module, prediction module, comparison module and analysis module; The acquisition module is used to acquire the real-time environmental parameters and the real-time power generation data of the photovoltaic panel assembly collected in real time; The prediction module is used to predict the photovoltaic panel components based on the real-time environmental parameters collected in real time and the constructed power prediction model to obtain theoretical power generation data; The comparison module is used to obtain the real-time power generation data of the photovoltaic panel assembly and compare it with the theoretical power generation data to obtain the comparison result and determine the abnormal photovoltaic panel assembly; The analysis module is used to analyze the image data of abnormal photovoltaic panel components acquired in real time to obtain abnormal results; The output module is used to output abnormal results.

Images