Distributed photovoltaic power quality monitoring analysis method, device, equipment and medium

Abstract

The invention provides a distributed photovoltaic power quality monitoring analysis method, device and equipment and a medium, and relates to the technical field of power quality evaluation. The method comprises the following steps: monitoring electric energy quality data of key nodes in a distributed photovoltaic power station, and preprocessing the data to obtain current feature data of a plurality of set feature indexes; based on the current feature data and the historical feature data within the set duration, an entropy weight method is adopted to distribute weights for all the set feature indexes so as to reflect the influence degree of different features on the electric energy quality; according to the weight and the current feature data, determining corrected feature data corresponding to the set feature index; and determining an electric energy quality analysis result of the distributed photovoltaic power station node according to the corrected feature data. According to the method, the actual operation state and potential problems of the power station can be reflected more accurately by comprehensively considering the plurality of set characteristic indexes and dynamically allocating the weights by using the entropy weight method, so that the accuracy of the electric energy quality analysis result is effectively improved.

Description

Technical Field

[0001] This application relates to the field of power quality assessment technology, and in particular to a method, device, equipment and medium for monitoring and analyzing the power quality of distributed photovoltaic power. Background Technology

[0002] Against the backdrop of modern energy transition, photovoltaic power generation, due to its green, efficient, and environmentally friendly characteristics, has gradually become an important alternative to non-renewable energy sources. However, with the widespread application of distributed photovoltaic power stations and the random connection of numerous power electronic devices and nonlinear loads, power quality problems such as harmonics, voltage fluctuations, and frequency deviations in the power grid are becoming increasingly prominent, posing a significant challenge to the stable operation and reliability of the power grid.

[0003] In related technologies, the total harmonic distortion (THD) of a photovoltaic (PV) power plant is typically determined by monitoring its harmonic content. The overall power quality of the PV power plant is then assessed based on the relationship between THD and the harmonic distortion threshold. However, this method suffers from low accuracy in obtaining the overall power quality. Summary of the Invention

[0004] This application provides a method, apparatus, equipment, and medium for monitoring and analyzing the power quality of distributed photovoltaic power, in order to improve the problem of low accuracy of power quality obtained through related technologies.

[0005] Firstly, this application provides a method for monitoring and analyzing the power quality of distributed photovoltaic power, including:

[0006] Monitor power quality data of key nodes in a distributed photovoltaic power station, including inverters, branch circuits, and combiner boxes;

[0007] The power quality data is preprocessed to obtain the current feature data of multiple set feature indicators corresponding to the power quality data;

[0008] Based on the current feature data and the historical feature data of multiple set feature indicators within a set time period, the entropy weight method is used to assign weights to each set feature indicator, and the weights corresponding to each set feature indicator are obtained. The weights represent the degree of influence of the corresponding set feature indicator on power quality.

[0009] For each set feature indicator, the corrected feature data corresponding to the set feature indicator is determined based on the weight of the set feature indicator and the current feature data.

[0010] Based on the corrected feature data corresponding to multiple set feature indicators, the power quality analysis results of the distributed photovoltaic power station nodes are determined.

[0011] In one possible implementation, the power quality analysis results of distributed photovoltaic power station nodes are determined based on the corrected feature data corresponding to multiple set feature indicators, including: inputting the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data into a pre-trained Long Short-Term Memory (LSTM) network model to obtain the power quality trend prediction results of distributed photovoltaic power station nodes; inputting the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data into a pre-trained random forest model to obtain the power quality anomaly diagnosis results of distributed photovoltaic power station nodes, the anomaly diagnosis results including the anomaly type and the probability of anomaly occurrence; and determining the power quality analysis results of distributed photovoltaic power station nodes based on the corrected feature data corresponding to multiple set feature indicators, the power quality trend prediction results, and the anomaly diagnosis results.

[0012] In one possible implementation, the power quality analysis results of distributed photovoltaic (PV) power station nodes are determined based on the corrected feature data corresponding to multiple set feature indicators, power quality trend prediction results, and anomaly diagnosis results. This includes: generating power quality scores for distributed PV power station nodes using a weighted scoring model based on the corrected feature data corresponding to the multiple set feature indicators; the power quality scores reflecting the health of the distributed PV power station nodes' operating status; classifying the power quality risk levels of distributed PV power station nodes based on the power quality scores and power quality trend prediction results, and generating power quality trend warning information; generating control suggestions for distributed PV power station nodes based on anomaly diagnosis results; and generating the power quality analysis results for distributed PV power station nodes based on the power quality scores, trend warning information, and control suggestions.

[0013] In one possible implementation, the distributed photovoltaic power quality monitoring and analysis method further includes: if the power quality score is within a first threshold range, then executing a first control strategy, which is used to monitor and provide early warning of the power quality of the distributed photovoltaic power station nodes; if the power quality score is within a second threshold range, then executing a second control strategy, which is used to trigger trend analysis to assess whether the power quality of the distributed photovoltaic power station nodes has further deteriorated; if the power quality score is within a third threshold range, then executing a third control strategy, which is used to trigger a control mechanism, including reducing the inverter output power, switching to a backup power supply, or adjusting the reactive power compensation device.

[0014] In one possible implementation, based on current feature data and historical feature data of multiple set feature indicators within a set time period, an entropy weight method is used to assign weights to each set feature indicator, resulting in weights corresponding to each set feature indicator. This includes: normalizing the current feature data based on historical feature data to obtain normalized feature data; determining the information entropy corresponding to each set feature indicator based on the normalized feature data; determining the initial weight of the set feature indicator based on the information entropy using the entropy weight method; and adjusting the initial weights of each set feature indicator in conjunction with environmental factors to obtain weights corresponding to each set feature indicator.

[0015] In one possible implementation, monitoring power quality data of key nodes in a distributed photovoltaic power station includes: collecting power quality data of key nodes in the distributed photovoltaic power station using a set sampling frequency, wherein the set sampling frequency is greater than or equal to 200kHz.

[0016] In one possible implementation, the power quality data includes voltage, current, frequency, harmonics, voltage fluctuations, and sag data. Multiple set characteristic indicators include THD, voltage fluctuation frequency, and sag duration. The power quality data is preprocessed to obtain current characteristic data corresponding to these multiple set characteristic indicators. This preprocessing includes: cleaning the power quality data to obtain cleaned power quality data; performing frequency domain analysis on the cleaned voltage and current time series data using Fast Fourier Transform (FFT) to obtain the current characteristic data corresponding to THD; and analyzing the cleaned voltage and cleaned sag data using wavelet transform to obtain the current characteristic data corresponding to voltage fluctuation frequency and sag duration, respectively.

[0017] Secondly, this application provides a distributed photovoltaic power quality monitoring and analysis device, comprising:

[0018] The data acquisition module is used to monitor the power quality data of key nodes in a distributed photovoltaic power station. Key nodes include inverters, branches, and combiner boxes.

[0019] The preprocessing module is used to preprocess the power quality data to obtain the current feature data of multiple set feature indicators corresponding to the power quality data;

[0020] The weight allocation module is used to assign weights to each set feature indicator based on the current feature data and the historical feature data of multiple set feature indicators within a set time period, using the entropy weight method, to obtain the weights corresponding to each set feature indicator. The weights represent the degree of influence of the corresponding set feature indicator on power quality. Furthermore, for each set feature indicator, the module determines the corrected feature data corresponding to the set feature indicator based on the weights corresponding to the set feature indicator and the current feature data.

[0021] The power quality analysis module is used to determine the power quality analysis results of distributed photovoltaic power station nodes based on the corrected feature data corresponding to multiple set feature indicators.

[0022] Thirdly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0023] Memory is used to store instructions executed by the computer;

[0024] A processor for executing computer-executable instructions stored in memory to implement any of the methods of the first aspect.

[0025] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed, are used to implement the method of any one of the first aspects.

[0026] Fifthly, this application provides a computer program product, including a computer program that, when executed, implements the method of any one of the first aspects.

[0027] The distributed photovoltaic (PV) power quality monitoring and analysis method, apparatus, equipment, and medium provided in this application monitor power quality data of key nodes in a distributed PV power station, including inverters, branches, and combiner boxes. The power quality data is preprocessed to obtain current characteristic data for multiple set characteristic indicators. Based on the current characteristic data and historical characteristic data of the multiple set characteristic indicators within a set time period, an entropy weight method is used to assign weights to each set characteristic indicator, obtaining the weights corresponding to each set characteristic indicator. The weights represent the degree of influence of the corresponding set characteristic indicator on power quality. For each set characteristic indicator, the corrected characteristic data corresponding to the set characteristic indicator is determined according to the weights and current characteristic data. Finally, the power quality analysis results of the distributed PV power station nodes are determined based on the corrected characteristic data corresponding to the multiple set characteristic indicators. In this process, by considering multiple set characteristic indicators and using the entropy weight method to assign weights to each set characteristic indicator, the weights can dynamically reflect the degree of influence of different characteristics on power quality, highlighting the characteristic indicators that mainly affect power quality. Compared with power quality analysis relying solely on a single characteristic indicator, this method can more accurately reflect the actual operating status and potential problems of distributed photovoltaic power stations, thereby effectively improving the accuracy of power quality analysis results. In addition, by monitoring the power quality data of key nodes in distributed photovoltaic power stations, compared with the traditional method of monitoring and analyzing the overall power quality of the grid-connected points of photovoltaic power stations through centralized monitoring equipment, this method can grasp the power quality status of each key node in real time, achieving independent monitoring of key nodes. This effectively overcomes the limitation of centralized monitoring not being able to refine the data to each key node, further improving the accuracy of power quality analysis results and meeting the needs of distributed photovoltaic power stations for refined management and control. At the same time, through real-time monitoring and analysis, abnormal power quality conditions at key nodes can be quickly identified, improving the anomaly response speed, which has important positive significance for ensuring the stable operation and safety and reliability of the power grid. Attached Figure Description

[0028] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0029] Figure 1 A flowchart illustrating a distributed photovoltaic power quality monitoring and analysis method provided as an exemplary embodiment of this application;

[0030] Figure 2 Another flowchart illustrating the distributed photovoltaic power quality monitoring and analysis method provided as an exemplary embodiment of this application;

[0031] Figure 3A schematic diagram of a distributed photovoltaic power quality monitoring and analysis device provided as an exemplary embodiment of this application;

[0032] Figure 4 Another structural schematic diagram of a distributed photovoltaic power quality monitoring and analysis device provided as an exemplary embodiment of this application;

[0033] Figure 5 A schematic diagram of the architecture of a distributed photovoltaic power quality monitoring and analysis device provided as an exemplary embodiment of this application;

[0034] Figure 6 A schematic diagram of the structure of an electronic device provided as an exemplary embodiment of this application.

[0035] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0036] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0037] The terms “first,” “second,” etc., used in the specification and claims of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, products, or apparatus.

[0038] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0039] In related technologies, power quality analysis relies solely on a single characteristic indicator, neglecting the effects of multi-parameter coupling such as harmonic superposition and voltage fluctuations. This results in an inaccurate reflection of the actual operating status and potential problems of distributed photovoltaic (PV) power plants, leading to low accuracy in power quality analysis results. Furthermore, traditional methods typically monitor and analyze the overall power quality of PV power plant grid connection points using centralized monitoring equipment. However, given the numerous, dispersed, and rapidly changing operating conditions of distributed PV power plants, this centralized monitoring approach has significant drawbacks. For example, it cannot achieve independent monitoring of key nodes, has slow anomaly response speed, and faces significant data transmission and processing pressures. This makes it difficult to meet the needs of real-time, refined power quality management of distributed PV power plants. These limitations further affect the accuracy of power quality analysis results, hindering the effective support for the stable operation and reliability of distributed PV power plants.

[0040] To address the aforementioned issues, this application provides a distributed photovoltaic (PV) power quality monitoring and analysis scheme. By comprehensively considering multiple set characteristic indicators, rather than relying solely on a single indicator, and employing the entropy weight method to assign weights to each indicator, the scheme quantifies the impact of different characteristics on power quality over different time periods. This highlights the key indicator influencing power quality, thereby more accurately reflecting the actual operating status and potential problems of the distributed PV power station and effectively improving the accuracy of power quality analysis results. Furthermore, intelligent monitoring devices are deployed at each key node, such as inverters, branches, or combiner boxes, enabling independent monitoring of these key nodes. This node-level monitoring can promptly capture power quality changes at each node, effectively improving anomaly response speed and reducing data transmission delays and processing pressure. This further enhances the accuracy of power quality analysis results for distributed PV power stations, which is of positive significance for supporting the stable operation and reliability of distributed PV systems.

[0041] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0042] Figure 1 A flowchart illustrating a distributed photovoltaic power quality monitoring and analysis method provided as an exemplary embodiment of this application is shown below. Figure 1 As shown, the distributed photovoltaic power quality monitoring and analysis method includes the following steps:

[0043] S101. Monitor the power quality data of key nodes in a distributed photovoltaic power station. Key nodes include inverters, branches, and combiner boxes.

[0044] A photovoltaic (PV) power plant is a power generation system that uses the photovoltaic effect to convert solar energy into electrical energy. It typically consists of PV modules, inverters, support systems, combiner boxes, transformers, and monitoring systems. Critical nodes refer to equipment and lines crucial for power quality monitoring and analysis, including but not limited to inverters, branch lines, and combiner boxes. An inverter is a device that converts the direct current (DC) generated by PV modules into alternating current (AC), and is the core component of a PV power generation system. Branch lines are electrical connection lines from PV strings to combiner boxes or inverters, typically including cables and connectors. Combiner boxes are devices that collect electrical energy from multiple PV strings and transmit it to the inverter; they typically have overcurrent protection and lightning protection functions.

[0045] For example, intelligent monitoring equipment is deployed at each key node to monitor the power quality data of key nodes in the distributed photovoltaic power station, achieving full coverage monitoring from "points to lines to surfaces".

[0046] S102. Preprocess the power quality data to obtain the current feature data of multiple set feature indicators corresponding to the power quality data.

[0047] In some embodiments, power quality data includes voltage, current, frequency, harmonics, voltage fluctuations, and sag data. Multiple set characteristic indicators include THD, voltage fluctuation frequency, and sag duration. Preprocessing the power quality data yields current characteristic data corresponding to these multiple set characteristic indicators. This preprocessing includes: cleaning the power quality data to obtain cleaned power quality data; performing frequency domain analysis on the cleaned voltage and current time series data using FFT to obtain the current characteristic data corresponding to THD; and analyzing the cleaned voltage and cleaned sag data using wavelet transform to obtain the current characteristic data corresponding to voltage fluctuation frequency and sag duration, respectively.

[0048] For example, power quality data is cleaned to remove outliers or missing values ​​caused by equipment failures or communication anomalies, resulting in cleaned power quality data. Based on the original voltage and current time series data contained in the power quality data, frequency domain analysis is performed using FFT to extract harmonic content, and the current feature data corresponding to THD is calculated based on the harmonic content. Wavelet transform is performed on the cleaned voltage data to identify the frequency of voltage fluctuation events, resulting in the current feature data corresponding to the voltage fluctuation frequency. Wavelet transform is performed on the cleaned sag data to identify the start and end times of sag events, calculate the sag duration, and obtain the current feature data corresponding to the sag duration.

[0049] S103. Based on the current feature data and the historical feature data of multiple set feature indicators within a set time period, the entropy weight method is used to assign weights to each set feature indicator to obtain the weights corresponding to each set feature indicator. The weights represent the degree of influence of the corresponding set feature indicator on power quality.

[0050] Among them, the entropy weight method is a weight allocation method based on information entropy, used to quantify the impact of various characteristic indicators on power quality. For example, by combining the current characteristic data and the historical characteristic data of multiple set characteristic indicators over a set period of time, such as 24 hours, the entropy weight method is used to assign weights to each set characteristic indicator, such as THD, voltage fluctuation frequency, and sag duration, to obtain the weights corresponding to THD, voltage fluctuation frequency, and sag duration, for example, THD weight is 0.3, voltage fluctuation frequency weight is 0.6, and sag duration weight is 0.1.

[0051] S104. For each set feature indicator, determine the corrected feature data corresponding to the set feature indicator based on the weight corresponding to the set feature indicator and the current feature data.

[0052] For example, the weight corresponding to the set feature index is multiplied by the current feature data to obtain the corrected feature data corresponding to the set feature index. For example, when the weight of THD is 0.3, the weight of voltage fluctuation frequency is 0.6, and the weight of sag duration is 0.1, the corrected feature data corresponding to the set feature index is determined by multiplying the weight corresponding to the set feature index by the current feature data.

[0053] S105. Based on the corrected feature data corresponding to multiple set feature indicators, determine the power quality analysis results of the distributed photovoltaic power station nodes.

[0054] For example, the corrected feature data corresponding to multiple set feature indicators are input into a pre-trained power quality analysis model. The power quality analysis model comprehensively considers the coupled effects of multiple set feature indicators, such as harmonic superposition and voltage fluctuations, to perform refined analysis and multi-dimensional evaluation of power quality characteristics. Based on the analysis and evaluation results, the power quality analysis results of the distributed photovoltaic power station nodes are determined.

[0055] The distributed photovoltaic power quality monitoring and analysis method provided in this application considers multiple set characteristic indicators and uses the entropy weight method to assign weights to each set characteristic indicator. This allows the weights to dynamically reflect the degree of influence of different characteristics on power quality, highlighting the characteristic indicators that mainly affect power quality. Compared with power quality analysis relying solely on a single characteristic indicator, this method can more accurately reflect the actual operating status and potential problems of distributed photovoltaic power stations, thereby effectively improving the accuracy of power quality analysis results. In addition, by monitoring the power quality data of key nodes in the distributed photovoltaic power station, compared with the traditional method of monitoring and analyzing the overall power quality of the photovoltaic power station grid connection point through centralized monitoring equipment, this method can grasp the power quality status of each key node in real time, achieving independent monitoring of key nodes. This effectively overcomes the limitation of centralized monitoring not being able to refine the data to each key node, further improving the accuracy of power quality analysis results and meeting the needs of distributed photovoltaic power stations for refined management and control. At the same time, through real-time monitoring and analysis, abnormal power quality conditions of key nodes can be quickly identified, improving the anomaly response speed, which has important positive significance for ensuring the stable operation and safety and reliability of the power grid.

[0056] In some embodiments, the power quality analysis results of distributed photovoltaic power station nodes are determined based on the corrected feature data corresponding to multiple set feature indicators, including: inputting the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data into a pre-trained LSTM model to obtain the power quality trend prediction results of distributed photovoltaic power station nodes; inputting the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data into a pre-trained random forest model to obtain the power quality anomaly diagnosis results of distributed photovoltaic power station nodes, the anomaly diagnosis results including the anomaly type and the probability of anomaly occurrence; and determining the power quality analysis results of distributed photovoltaic power station nodes based on the corrected feature data corresponding to multiple set feature indicators, the power quality trend prediction results, and the anomaly diagnosis results.

[0057] For example, the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data of the past 24 hours are input into a pre-trained LSTM model to obtain the power quality trend prediction results of distributed photovoltaic power station nodes. For example, the harmonic content is expected to increase by 10% in the next 10 minutes, the voltage fluctuation frequency is expected to increase by 5 times / hour in the next 10 minutes, or the sag duration is expected to increase by 0.1 seconds in the next 10 minutes. At the same time, the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data of the past 24 hours are input into a pre-trained random forest model. This model uses the random forest algorithm to classify the input feature data and identifies the potential causes of power quality anomalies based on feature combination rules constructed from historical abnormal events (e.g., "diagnose inverter fault when voltage drops and light intensity decreases"). For example, if THD suddenly increases from 3% to 8% and the load current fluctuates frequently, the diagnosis is that the load surge caused harmonic anomalies, with an 85% probability of occurrence; if the voltage drops instantaneously by 15% and the light intensity drops sharply, the diagnosis is that the inverter switching is abnormal, with a 90% probability of occurrence; if the frequency deviation is ±0.6Hz and the reactive power fluctuates drastically, the diagnosis is that the reactive power compensation device is faulty, with a 95% probability of occurrence, and so on. Furthermore, based on the corrected characteristic data, power quality trend prediction results, and anomaly diagnosis results corresponding to multiple set characteristic indicators, the power quality analysis results of the distributed photovoltaic power station nodes are determined.

[0058] It is understandable that the end of the model training process is similar to that in related technologies. It can be considered that the LSTM model or random forest model is completed after the set number of training iterations has been reached, or the LSTM model or random forest model is considered to be completed when the loss value is less than a set threshold.

[0059] This application embodiment utilizes an LSTM model to perform complex feature analysis on power quality, enabling accurate prediction of power quality trends. Simultaneously, a random forest model is used to accurately diagnose power quality anomalies, including identifying anomaly types and calculating the probability of anomaly occurrence. Combined with corrected feature data, a comprehensive assessment of the power quality status of distributed photovoltaic power station nodes can be achieved, providing effective data support for developing improvement measures and optimization strategies. This not only improves the accuracy and timeliness of power quality analysis but also provides a solid technical guarantee for the efficient operation and management of distributed photovoltaic power stations.

[0060] Based on the above embodiments, in some embodiments, the power quality analysis results of distributed photovoltaic (PV) power station nodes are determined based on the corrected feature data corresponding to multiple set feature indicators, power quality trend prediction results, and anomaly diagnosis results. This includes: generating power quality scores for distributed PV power station nodes using a weighted scoring model based on the corrected feature data corresponding to multiple set feature indicators; the power quality scores reflecting the health of the distributed PV power station nodes' operating status; classifying the power quality risk levels of distributed PV power station nodes based on the power quality scores and power quality trend prediction results, and generating power quality trend warning information; generating control suggestions for distributed PV power station nodes based on anomaly diagnosis results; and generating power quality analysis results for distributed PV power station nodes based on the power quality scores, trend warning information, and control suggestions.

[0061] For example, assuming the corrected feature data for THD is 0.15, the corrected feature data for voltage fluctuation frequency is 0.3, and the corrected feature data for sag duration is 0.05; using a weighted scoring model, the corrected feature data corresponding to multiple set feature indicators are summed to obtain an initial value of power quality score, for example, 0.5. This initial value is then converted to a percentage, resulting in a power quality score of, for example, 50 for the distributed photovoltaic power station node. Correspondingly, according to pre-set risk level classification rules, for example, a risk level of 80-100 points is normal, 60-79 points is slightly abnormal, and 0-59 points is severely abnormal, the risk level of power quality for the distributed photovoltaic power station node is classified based on the power quality score. A power quality score of 50 corresponds to a risk level of "severe anomaly." The power quality score and risk level reflect the current operating status of distributed photovoltaic (PV) power station nodes. Based on power quality trend prediction results, trend warning information is generated, such as a predicted 10% increase in harmonic content, a predicted increase of 5 voltage fluctuations per hour, or a predicted increase of 0.1 seconds in sag duration within the next 10 minutes. Based on anomaly diagnosis results, control recommendations for distributed PV power station nodes are generated, such as reducing inverter output power by 20%, checking reactive power compensation devices, or switching to backup power. Furthermore, based on the power quality score, trend warning information, and control recommendations, power quality analysis results for distributed PV power station nodes are generated, including health scores, risk levels, trend warnings, and control recommendations.

[0062] It should be noted that the above-mentioned pre-set risk level classification rules are only an example. In actual applications, they can be flexibly adjusted according to actual needs. There are no restrictions on the setting of risk level classification rules here.

[0063] In this embodiment, a power quality score is generated by combining steady-state parameter THD, transient event voltage fluctuation frequency, and long-term trend sag duration. Based on the power quality score, risk level, trend warning information, and control suggestions, power quality is analyzed from multiple dimensions, which can comprehensively reflect the power quality status of distributed photovoltaic power station nodes and further improve the accuracy of power quality analysis results.

[0064] In some embodiments, the distributed photovoltaic power quality monitoring and analysis method further includes: if the power quality score is within a first threshold range, then executing a first control strategy, the first control strategy being used to monitor and provide early warning of the power quality of the distributed photovoltaic power station nodes; if the power quality score is within a second threshold range, then executing a second control strategy, the second control strategy being used to trigger trend analysis to assess whether the power quality of the distributed photovoltaic power station nodes has further deteriorated; if the power quality score is within a third threshold range, then executing a third control strategy, the third control strategy being used to trigger a control mechanism, the control mechanism including reducing the inverter output power, switching to a backup power supply, or adjusting the reactive power compensation device.

[0065] For example, assuming the first threshold range is 80-100 points, the second threshold range is 60-79 points, and the third threshold range is 0-59 points; correspondingly, if the power quality score is between 80-100 points, the first control strategy is executed, which is to monitor and issue early warnings for the power quality of the distributed photovoltaic power station nodes; if the power quality score is between 60-79 points, the second control strategy is executed, which triggers trend analysis to assess whether the power quality will deteriorate further, for example, predicting future power quality trends based on LSTM models, generating trend warning information (such as the harmonic content is expected to increase by 5% in the next 10 minutes, and the voltage fluctuation frequency is expected to increase by 3 times / hour), and generating a soft alarm to prompt maintenance personnel to intervene and check; if the power quality score is between 0-59 points, the third control strategy is executed, which triggers the control mechanism, such as reducing the inverter output power by 20%, switching to backup power, or adjusting the reactive power compensation device, and generating an emergency alarm such as "Power quality is seriously abnormal, control measures have been implemented, please handle immediately".

[0066] In this embodiment, by setting different threshold ranges and implementing corresponding control strategies, differentiated response measures can be taken according to the severity of power quality anomalies. This hierarchical control strategy not only improves response efficiency but also avoids the problems of over-intervention or under-response, ensuring more precise and efficient power quality management. In addition, the implementation of the hierarchical control strategy can effectively reduce operation and maintenance costs and effectively reduce equipment damage or system downtime caused by power quality problems. At the same time, through the combination of real-time monitoring, trend analysis, and proactive control, power quality problems in distributed photovoltaic power station nodes can be detected and addressed in a timely manner, providing a reliable guarantee for the long-term stable operation of the power station.

[0067] In some embodiments, based on current feature data and historical feature data of multiple set feature indicators within a set time period, an entropy weight method is used to assign weights to each set feature indicator, resulting in weights corresponding to each set feature indicator. This includes: normalizing the current feature data based on historical feature data to obtain normalized feature data; determining the information entropy corresponding to each set feature indicator based on the normalized feature data; determining the initial weight of the set feature indicator based on the information entropy using the entropy weight method; and adjusting the initial weights corresponding to each set feature indicator in conjunction with environmental factors to obtain weights corresponding to each set feature indicator.

[0068] For example, based on historical feature data within a set time period, the current feature data is normalized. This involves converting feature data with different dimensions or value ranges corresponding to multiple set feature indicators, such as THD, voltage fluctuation frequency, and sag duration, into the [0, 1] interval to eliminate dimensional differences between different feature indicators. For each set feature indicator, the information entropy of its corresponding normalized feature data is calculated. Information entropy reflects the dispersion of feature data; the higher the dispersion, the greater the information entropy. Using the entropy weight method, the initial weight of each set feature indicator is determined based on the calculated information entropy. The smaller the information entropy of a feature indicator, the greater its weight, indicating a more significant impact of that feature on power quality. Furthermore, combined with the current environmental data... Environmental factors (such as light intensity or temperature) are considered when adjusting the initial weights of each set characteristic indicator. For example, during periods of high light intensity, such as midday on a sunny day, the large output of photovoltaic power and the full-load operation of the inverter can easily lead to harmonic distortion (THD). Therefore, the weight corresponding to THD is adjusted from 0.15 to 0.3. Conversely, during periods of low light intensity, such as evening on a cloudy day, frequent fluctuations in photovoltaic output can cause voltage spikes or dips. Therefore, the weight corresponding to the frequency of voltage fluctuations is adjusted from 0.1 to 0.35, and the weight corresponding to the duration of voltage dips is adjusted from 0.1 to 0.25. After these adjustments, the final weights corresponding to each set characteristic indicator are obtained. These weights are used for subsequent power quality analysis and decision support, ensuring that the power quality analysis results are more targeted and accurate.

[0069] In this embodiment, the entropy weight method is used to assign initial weights to each set characteristic indicator. This method can objectively reflect the degree of influence of each indicator on power quality based on information entropy, avoiding the bias that may be caused by subjective weighting, and significantly improving the objectivity and accuracy of power quality assessment. On this basis, the initial weights are dynamically adjusted in combination with environmental factors to further optimize the weight allocation results, making the weight allocation more in line with the actual operating environment, enhancing the pertinence and practicality of power quality assessment, and thus further improving the scientificity and accuracy of the assessment.

[0070] Considering that traditional centralized monitoring equipment in related technologies has a low sampling frequency, such as less than or equal to ≤10 kHz, resulting in high data latency, such as delays on the order of seconds or even minutes, it is impossible to accurately capture transient events such as microsecond-level voltage flicker, thus affecting the accuracy of power quality analysis results. Therefore, in some embodiments, monitoring power quality data of key nodes in a distributed photovoltaic power station includes: collecting power quality data of key nodes in the distributed photovoltaic power station using a set sampling frequency, with the set sampling frequency being greater than or equal to 200 kHz.

[0071] For example, signals such as voltage, current, frequency, harmonics, voltage fluctuations, and sag data of key nodes are continuously sampled at a sampling frequency of 200 kHz, thereby ensuring high-precision capture of high-order harmonics (such as harmonics above the 50th order) and transient events (such as microsecond-level voltage fluctuations).

[0072] Figure 2 Another schematic flowchart of a distributed photovoltaic power quality monitoring and analysis method provided as an exemplary embodiment of this application is shown. Figure 2 As shown, the distributed photovoltaic power quality monitoring and analysis method includes the following steps:

[0073] S201. Monitor the power quality data of key nodes in a distributed photovoltaic power station. Key nodes include inverters, branches, and combiner boxes.

[0074] For example, intelligent monitoring equipment is deployed at each key node to monitor the power quality data of key nodes in the distributed photovoltaic power station, achieving full coverage monitoring from "points to lines to surfaces".

[0075] S202. Preprocess the power quality data to obtain the current feature data of multiple set feature indicators corresponding to the power quality data.

[0076] The power quality data includes voltage, current, frequency, harmonics, voltage fluctuations, and sag data. Several set characteristic indicators include THD, voltage fluctuation frequency, and sag duration.

[0077] For example, power quality data is cleaned to obtain cleaned power quality data; FFT is used to perform frequency domain analysis on the cleaned voltage time series data and current time series data to obtain the current feature data corresponding to THD; wavelet transform is used to analyze the cleaned voltage and cleaned sag data respectively to obtain the current feature data corresponding to voltage fluctuation frequency and sag duration.

[0078] S203. Based on the current feature data and the historical feature data of multiple set feature indicators within a set time period, the entropy weight method is used to assign weights to each set feature indicator to obtain the weights corresponding to each set feature indicator.

[0079] The weights represent the degree of influence of the corresponding set characteristic indicators on power quality.

[0080] Specifically, based on historical feature data, the current feature data is normalized to obtain normalized feature data; for each set feature indicator, the information entropy corresponding to the set feature indicator is determined according to the normalized feature data corresponding to the set feature indicator; using the entropy weight method, the initial weight of the set feature indicator is determined according to the information entropy; and combined with environmental factors, the initial weights corresponding to each set feature indicator are adjusted to obtain the weights corresponding to each set feature indicator.

[0081] S204. For each set feature indicator, determine the corrected feature data corresponding to the set feature indicator based on the weight corresponding to the set feature indicator and the current feature data.

[0082] For example, the weight corresponding to the set feature index is multiplied by the current feature data to obtain the corrected feature data corresponding to the set feature index. For example, when the weight of THD is 0.3, the weight of voltage fluctuation frequency is 0.6, and the weight of sag duration is 0.1, the corrected feature data corresponding to the set feature index is determined by multiplying the weight corresponding to the set feature index by the current feature data.

[0083] S205. Input the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data into the pre-trained LSTM model and the pre-trained random forest model to obtain the power quality trend prediction results and power quality anomaly diagnosis results of distributed photovoltaic power station nodes.

[0084] For example, the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data of the past 24 hours are input into a pre-trained LSTM model to obtain the power quality trend prediction results of distributed photovoltaic power station nodes. For example, the harmonic content is expected to increase by 10% in the next 10 minutes, the voltage fluctuation frequency is expected to increase by 5 times / hour in the next 10 minutes, or the sag duration is expected to increase by 0.1 seconds in the next 10 minutes. At the same time, the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data of the past 24 hours are input into a pre-trained random forest model to identify potential causes of power quality anomalies. For example, if THD suddenly increases from 3% to 8% and the load current fluctuates frequently, the diagnosis is that the load impact caused the harmonic anomaly, with a probability of 85%; if the voltage drops instantaneously by 15% and the light intensity drops sharply, the diagnosis is that the inverter switching is abnormal, with a probability of 90%; if the frequency deviation is ±0.6Hz and the reactive power fluctuates drastically, the diagnosis is that the reactive power compensation device is faulty, with a probability of 95%, etc.

[0085] S206. Based on the corrected feature data corresponding to multiple set feature indicators, a weighted scoring model is used to generate the power quality score of the distributed photovoltaic power station node.

[0086] The power quality score reflects the health of the distributed photovoltaic (PV) power station nodes. For example, assuming the corrected feature data for THD is 0.15, the corrected feature data for voltage fluctuation frequency is 0.3, and the corrected feature data for sag duration is 0.05, a weighted scoring model is used to sum the corrected feature data corresponding to each of the multiple set feature indicators to obtain an initial value for the power quality score, for example, 0.5. This initial value is then converted to a percentage, resulting in a power quality score of, for example, 50 points for the distributed PV power station node.

[0087] S207. Based on the power quality score and power quality trend prediction results, classify the power quality risk level of distributed photovoltaic power station nodes and generate power quality trend warning information.

[0088] For example, according to pre-set classification rules, such as a risk level of 80-100 points being normal, a risk level of 60-79 points being slightly abnormal, and a risk level of 0-59 points being severely abnormal; the risk level of the power quality of distributed photovoltaic power station nodes is classified according to the power quality score, for example, a risk level of "severely abnormal" is corresponding to a power quality score of 50 points.

[0089] S208. Based on the anomaly diagnosis results, generate control recommendations for nodes in the distributed photovoltaic power station.

[0090] The control recommendations include, but are not limited to, reducing the inverter output power by 20%, checking the reactive power compensation device, or switching to a backup power source.

[0091] S209. Based on power quality score, trend warning information and control suggestions, generate power quality analysis results for distributed photovoltaic power station nodes.

[0092] The power quality analysis results include health scores, risk levels, trend warnings, and control recommendations.

[0093] S210. Based on the target threshold range corresponding to the power quality fraction, execute the control strategy corresponding to the target threshold range.

[0094] For example, if the power quality score is within a first threshold range, a first control strategy is executed. The first control strategy is used to monitor and provide early warning of the power quality of the distributed photovoltaic power station nodes. If the power quality score is within a second threshold range, a second control strategy is executed. The second control strategy is used to trigger trend analysis to assess whether the power quality of the distributed photovoltaic power station nodes has further deteriorated. If the power quality score is within a third threshold range, a third control strategy is executed. The third control strategy is used to trigger a control mechanism, which includes reducing the inverter output power, switching to a backup power supply, or adjusting the reactive power compensation device.

[0095] In summary, this application has at least the following advantages:

[0096] First, by considering multiple set characteristic indicators and employing the entropy weight method to assign weights to each indicator, the weights can dynamically reflect the degree of influence of different characteristics on power quality, highlighting the main characteristic indicators affecting power quality. Compared to relying solely on a single characteristic indicator for power quality analysis, this method can more accurately reflect the actual operating status and potential problems of distributed photovoltaic power stations, thereby effectively improving the accuracy of power quality analysis results. Second, by monitoring the power quality data of key nodes in distributed photovoltaic power stations, compared to the traditional method of monitoring and analyzing the overall power quality of the grid-connected points of photovoltaic power stations through centralized monitoring equipment, this method can grasp the power quality status of each key node in real time, achieving independent monitoring of key nodes. This effectively overcomes the limitations of centralized monitoring in not being able to refine the data to each key node, further improving the accuracy of power quality analysis results and meeting the needs of distributed photovoltaic power stations for refined management and control. Simultaneously, through real-time monitoring and analysis, abnormal power quality conditions at key nodes can be quickly identified, improving the anomaly response speed and playing a significant positive role in ensuring the stable operation and reliability of the power grid.

[0097] Second, by utilizing the LSTM model to perform complex feature analysis on power quality, accurate prediction of power quality trends can be achieved. Simultaneously, the random forest model is used to accurately diagnose power quality anomalies, including identifying anomaly types and calculating the probability of anomaly occurrence. Combined with corrected feature data, a comprehensive assessment of the power quality status of distributed photovoltaic power station nodes can be conducted, providing effective data support for formulating improvement measures and optimization strategies. This not only improves the accuracy and timeliness of power quality analysis but also provides a solid technical guarantee for the efficient operation and management of distributed photovoltaic power stations.

[0098] Third, by setting different threshold ranges and implementing corresponding control strategies, differentiated response measures can be taken according to the severity of power quality anomalies. This hierarchical control strategy not only improves response efficiency but also avoids the problems of over-intervention or under-response, ensuring more precise and efficient power quality management. In addition, the implementation of the hierarchical control strategy can effectively reduce operation and maintenance costs and effectively reduce equipment damage or system downtime caused by power quality problems. At the same time, through the combination of real-time monitoring, trend analysis, and proactive control, power quality problems in distributed photovoltaic power station nodes can be detected and addressed in a timely manner, providing a reliable guarantee for the long-term stable operation of the power station.

[0099] Fourth, by assigning initial weights to each set characteristic indicator using the entropy weight method, the impact of each indicator on power quality can be objectively reflected based on information entropy, avoiding the bias that may be caused by subjective weighting, and significantly improving the objectivity and accuracy of power quality assessment. On this basis, the initial weights are dynamically adjusted in combination with environmental factors to further optimize the weight allocation results, making the weight allocation more in line with the actual operating environment, enhancing the pertinence and practicality of power quality assessment, and thus further improving the scientificity and accuracy of the assessment.

[0100] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0101] Figure 3 A schematic diagram of a distributed photovoltaic power quality monitoring and analysis device provided for an exemplary embodiment of this application. The distributed photovoltaic power quality monitoring and analysis device 30 provided in this embodiment includes a data acquisition module 31, a preprocessing module 32, a weight allocation module 33, and a power quality analysis module 34, wherein:

[0102] The data acquisition module 31 is used to monitor the power quality data of key nodes in the distributed photovoltaic power station. The key nodes include inverters, branches and combiner boxes.

[0103] Preprocessing module 32 is used to preprocess the power quality data to obtain the current feature data of multiple set feature indicators corresponding to the power quality data;

[0104] The weight allocation module 33 is used to assign weights to each set feature indicator based on the current feature data and the historical feature data of multiple set feature indicators within a set time period, using the entropy weight method, to obtain the weights corresponding to each set feature indicator. The weights represent the degree of influence of the corresponding set feature indicator on power quality. The module also determines the corrected feature data corresponding to each set feature indicator based on the weights corresponding to the set feature indicator and the current feature data.

[0105] The power quality analysis module 34 is used to determine the power quality analysis results of the distributed photovoltaic power station nodes based on the corrected feature data corresponding to multiple set feature indicators.

[0106] In one possible implementation, the power quality analysis module 34 can be specifically used to: input the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data into a pre-trained LSTM model to obtain the power quality trend prediction result of the distributed photovoltaic power station node; input the corrected feature data corresponding to multiple set feature indicators and the corrected feature data corresponding to historical feature data into a pre-trained random forest model to obtain the power quality anomaly diagnosis result of the distributed photovoltaic power station node, the anomaly diagnosis result including the anomaly type and the probability of anomaly occurrence; and determine the power quality analysis result of the distributed photovoltaic power station node based on the corrected feature data corresponding to multiple set feature indicators, the power quality trend prediction result, and the anomaly diagnosis result.

[0107] In one possible implementation, the power quality analysis module 34 can also be used to: generate power quality scores for distributed photovoltaic power station nodes based on corrected feature data corresponding to multiple set feature indicators, using a weighted scoring model; the power quality scores reflect the health of the operating status of the distributed photovoltaic power station nodes; classify the risk level of power quality for distributed photovoltaic power station nodes according to the power quality scores and power quality trend prediction results, and generate power quality trend warning information; generate control suggestions for distributed photovoltaic power station nodes based on anomaly diagnosis results; and generate power quality analysis results for distributed photovoltaic power station nodes based on power quality scores, trend warning information, and control suggestions.

[0108] In one possible implementation, the preprocessing module 32 can be specifically used to: normalize the current feature data based on historical feature data to obtain normalized feature data; determine the information entropy corresponding to each set feature indicator based on the normalized feature data corresponding to the set feature indicator; determine the initial weight of the set feature indicator based on the information entropy using the entropy weight method; and adjust the initial weight of each set feature indicator in combination with environmental factors to obtain the weight corresponding to each set feature indicator.

[0109] In one possible implementation, the data acquisition module 31 can be specifically used to: acquire power quality data of key nodes in a distributed photovoltaic power station using a set sampling frequency, wherein the set sampling frequency is greater than or equal to 200kHz.

[0110] In one possible implementation, the power quality data includes voltage, current, frequency, harmonics, voltage fluctuations, and sag data. Multiple set characteristic indicators include THD, voltage fluctuation frequency, and sag duration. The preprocessing module 32 can also be used to: clean the power quality data to obtain cleaned power quality data; perform frequency domain analysis on the cleaned voltage and current time series data using Fast Fourier Transform (FFT) to obtain the current characteristic data corresponding to THD; and analyze the cleaned voltage and cleaned sag data using wavelet transform to obtain the current characteristic data corresponding to voltage fluctuation frequency and sag duration, respectively.

[0111] Figure 4 Another structural schematic diagram of the distributed photovoltaic power quality monitoring and analysis device provided for an exemplary embodiment of this application. The distributed photovoltaic power quality monitoring and analysis device 30 provided in this embodiment further includes an early warning feedback module 35, wherein:

[0112] The early warning feedback module 35 can be specifically used to: execute a first control strategy when the power quality score is within a first threshold range, the first control strategy being used to monitor and provide early warning of the power quality of the distributed photovoltaic power station nodes; execute a second control strategy when the power quality score is within a second threshold range, the second control strategy being used to trigger trend analysis to assess whether the power quality of the distributed photovoltaic power station nodes has further deteriorated; and execute a third control strategy when the power quality score is within a third threshold range, the third control strategy being used to trigger a control mechanism, the control mechanism including reducing the inverter output power, switching to a backup power supply, or adjusting the reactive power compensation device.

[0113] For example, Figure 5 A schematic diagram of the architecture of a distributed photovoltaic power quality monitoring and analysis device provided as an exemplary embodiment of this application. Figure 5As shown, the architecture comprises a monitoring layer, an analysis layer, and a feedback control layer. The monitoring layer, serving as the foundational sensing layer, collects real-time power quality data from key nodes in the distributed photovoltaic power station via a data acquisition module. This power quality data includes voltage, current, frequency, harmonics, voltage fluctuations, and sag data. A preprocessing module cleans the data and extracts feature data, then transmits the preprocessed data to the analysis layer in the form of standardized data packets, providing fundamental data support for further analysis. Accordingly, after receiving the feature data uploaded by the monitoring layer, the analysis layer performs in-depth analysis using algorithm models. Specifically, this includes: assigning weights to each set feature indicator using the entropy weight method; combining LSTM and random forest models to perform power quality trend prediction and anomaly diagnosis, obtaining power quality trend prediction results and anomaly diagnosis results; generating power quality scores for distributed photovoltaic power station nodes based on the corrected feature data corresponding to multiple set feature indicators using a weighted scoring model; classifying the power quality risk level of distributed photovoltaic power station nodes based on the power quality scores and power quality trend prediction results; generating power quality trend warning information; generating control suggestions for distributed photovoltaic power station nodes based on the anomaly diagnosis results; further, generating power quality analysis results for distributed photovoltaic power station nodes based on the power quality scores, trend warning information, and control suggestions; and sending the power quality analysis results to the feedback control layer. Correspondingly, after obtaining the power quality analysis results sent by the analysis layer, the feedback control layer executes the control strategy corresponding to the target threshold range based on the power quality score. In particular, when the power quality exceeds the limit or is abnormal, it triggers an early warning, records the details of the abnormal event, and feeds it back to the operation and maintenance personnel so that timely control measures can be taken.

[0114] The distributed photovoltaic power quality monitoring and analysis device provided in this application embodiment can execute the technical solution shown in the above-described distributed photovoltaic power quality monitoring and analysis method embodiment. Its implementation principle and beneficial effects are similar, and will not be repeated here.

[0115] It should be noted that the division of the various modules in the above device is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software via processing elements; they can be fully implemented in hardware; or some modules can be implemented by processing elements calling software, while others are implemented in hardware. For example, the power quality analysis module can be a separate processing element, or it can be integrated into a chip in the above device. Alternatively, it can be stored as program code in the device's memory, and its functions can be called and executed by a processing element. The implementation of other modules is similar. Moreover, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed through integrated logic circuits in the hardware of the processor element or through software instructions.

[0116] For example, these modules can be one or more integrated circuits configured to implement the above methods, such as one or more Application Specific Integrated Circuits (ASICs), one or more Digital Signal Processors (DSPs), or one or more Field Programmable Gate Arrays (FPGAs). As another example, when a module is implemented using processing element scheduler code, the processing element can be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. Furthermore, these modules can be integrated together as a System-On-a-Chip (SOC).

[0117] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is 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, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) 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 integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Video Discs, DVDs), or semiconductor media (e.g., solid-state disks (SSDs)).

[0118] Figure 6 A schematic diagram of the structure of an electronic device provided as an exemplary embodiment of this application. For example... Figure 6 As shown, the electronic device 60 in this embodiment includes:

[0119] At least one processor 61; and a memory 62 communicatively connected to the at least one processor;

[0120] The memory 62 stores instructions that can be executed by at least one processor 61 to cause the electronic device to perform the method as described in any of the above embodiments.

[0121] Alternatively, the memory 62 can be either standalone or integrated with the processor 61.

[0122] The memory 62 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device.

[0123] The processor 61 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. Specifically, when implementing the distributed photovoltaic power quality monitoring and analysis method described in the foregoing method embodiments, the electronic device may be, for example, an electronic device with processing capabilities such as a server.

[0124] Optionally, the electronic device may also include a communication interface 63. In specific implementations, if the communication interface 63, memory 62, and processor 61 are implemented independently, they can be interconnected via a bus to complete communication. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc., but this does not imply that there is only one bus or one type of bus.

[0125] Optionally, in a specific implementation, if the communication interface 63, memory 62, and processor 61 are integrated on a single chip, then the communication interface 63, memory 62, and processor 61 can communicate through an internal interface.

[0126] The implementation principle and technical effects of the electronic device provided in this embodiment can be found in the foregoing embodiments, and will not be repeated here.

[0127] This application also provides a computer-readable storage medium storing computer-executable instructions. When the computer-executable instructions are executed, they are used to implement the method steps as described in the above method embodiments. The specific implementation methods and technical effects are similar and will not be repeated here.

[0128] The aforementioned computer-readable storage media can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0129] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an application-specific integrated circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in a distributed photovoltaic power quality monitoring and analysis device.

[0130] This application also provides a computer program product, including a computer program, which, when executed, implements the method steps as described in the above method embodiments. The specific implementation and technical effects are similar and will not be repeated here.

[0131] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0132] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0133] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for monitoring and analyzing the power quality of distributed photovoltaic power, characterized in that, include: Monitor power quality data of key nodes in a distributed photovoltaic power station, wherein the key nodes include inverters, branches and combiner boxes; The power quality data is preprocessed to obtain the current feature data of multiple set feature indicators corresponding to the power quality data; Based on the current feature data and the historical feature data of the multiple set feature indicators within a set time period, the entropy weight method is used to assign weights to each set feature indicator to obtain the weights corresponding to each set feature indicator. The weights represent the degree of influence of the corresponding set feature indicator on power quality. For each of the defined feature indicators, the corrected feature data corresponding to the defined feature indicator is determined based on the weight corresponding to the defined feature indicator and the current feature data. Based on the corrected feature data corresponding to the multiple set feature indicators, the power quality analysis results of the distributed photovoltaic power station nodes are determined.

2. The distributed photovoltaic power quality monitoring and analysis method according to claim 1, characterized in that, The step of determining the power quality analysis results of distributed photovoltaic power station nodes based on the corrected feature data corresponding to the multiple set feature indicators includes: The corrected feature data corresponding to the multiple set feature indicators and the corrected feature data corresponding to the historical feature data are input into a pre-trained long short-term memory neural network model to obtain the power quality trend prediction results of the distributed photovoltaic power station nodes. The corrected feature data corresponding to the multiple set feature indicators and the corrected feature data corresponding to the historical feature data are input into a pre-trained random forest model to obtain the abnormal power quality diagnosis results of the distributed photovoltaic power station nodes. The abnormal diagnosis results include the abnormality type and the probability of the abnormality. Based on the corrected feature data corresponding to the multiple set feature indicators, the power quality trend prediction results, and the anomaly diagnosis results, the power quality analysis results of the distributed photovoltaic power station nodes are determined.

3. The distributed photovoltaic power quality monitoring and analysis method according to claim 2, characterized in that, The determination of the power quality analysis results for the distributed photovoltaic power station nodes based on the corrected feature data corresponding to the multiple set feature indicators, the power quality trend prediction results, and the anomaly diagnosis results includes: Based on the corrected feature data corresponding to the multiple set feature indicators, a weighted scoring model is used to generate the power quality score of the distributed photovoltaic power station node. The power quality score reflects the health of the operating status of the distributed photovoltaic power station node. Based on the power quality score and the power quality trend prediction results, the risk level of power quality of the distributed photovoltaic power station nodes is classified, and power quality trend early warning information is generated. Based on the anomaly diagnosis results, control recommendations are generated for the nodes of the distributed photovoltaic power station; Based on the power quality score, the trend warning information, and the control suggestions, the power quality analysis results of the distributed photovoltaic power station nodes are generated.

4. The distributed photovoltaic power quality monitoring and analysis method according to claim 3, characterized in that, Also includes: If the power quality score is within the first threshold range, a first control strategy is executed. The first control strategy is used to monitor and provide early warning of the power quality of the distributed photovoltaic power station nodes. If the power quality score is within the second threshold range, a second control strategy is executed. The second control strategy is used to trigger trend analysis to assess whether the power quality of the distributed photovoltaic power station node has further deteriorated. If the power quality fraction is within the third threshold range, a third control strategy is executed. The third control strategy is used to trigger a control mechanism, which includes reducing the inverter output power, switching to a backup power supply, or adjusting the reactive power compensation device.

5. The method for monitoring and analyzing the power quality of distributed photovoltaic power according to any one of claims 1 to 4, characterized in that, The process involves assigning weights to each set feature indicator based on the current feature data and historical feature data of the multiple set feature indicators within a set time period, using the entropy weight method to obtain the weights corresponding to each set feature indicator, including: Based on the historical feature data, the current feature data is normalized to obtain normalized feature data. For each of the defined feature indicators, the information entropy corresponding to the defined feature indicator is determined based on the normalized feature data corresponding to the defined feature indicator. Using the entropy weight method, the initial weights of the set feature indicators are determined based on the information entropy; By taking into account environmental factors, the initial weights corresponding to each of the set feature indicators are adjusted to obtain the weights corresponding to each of the set feature indicators.

6. The method for monitoring and analyzing the power quality of distributed photovoltaic power according to any one of claims 1 to 4, characterized in that, The power quality data monitored at key nodes in the distributed photovoltaic power station includes: Power quality data of key nodes in the distributed photovoltaic power station are collected using a set sampling frequency, wherein the set sampling frequency is greater than or equal to 200kHz.

7. The method for monitoring and analyzing the power quality of distributed photovoltaic power according to any one of claims 1 to 4, characterized in that, The power quality data includes voltage, current, frequency, harmonics, voltage fluctuations, and sag data. The multiple set characteristic indicators include total harmonic distortion (THD), voltage fluctuation frequency, and sag duration. The power quality data is preprocessed to obtain the current characteristic data corresponding to the multiple set characteristic indicators, including: The power quality data is cleaned to obtain cleaned power quality data; Fast Fourier Transform was used to perform frequency domain analysis on the cleaned voltage and current time series data to obtain the current characteristic data corresponding to the total harmonic distortion rate. Wavelet transform was used to analyze the voltage and sag data after cleaning, respectively, to obtain the current feature data corresponding to the voltage fluctuation frequency and sag duration.

8. A distributed photovoltaic power quality monitoring and analysis device, characterized in that, include: The data acquisition module is used to monitor the power quality data of key nodes in a distributed photovoltaic power station, including inverters, branches, and combiner boxes. The preprocessing module is used to preprocess the power quality data to obtain the current feature data of the power quality data corresponding to multiple set feature indicators. The weight allocation module is used to assign weights to each of the set feature indicators based on the current feature data and the historical feature data of the multiple set feature indicators within a set time period, using the entropy weight method, to obtain the weights corresponding to each of the set feature indicators, wherein the weights represent the degree of influence of the corresponding set feature indicator on power quality; and, for each set feature indicator, to determine the corrected feature data corresponding to the set feature indicator based on the weights corresponding to the set feature indicator and the current feature data. The power quality analysis module is used to determine the power quality analysis results of the distributed photovoltaic power station nodes based on the corrected feature data corresponding to the multiple set feature indicators.

9. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory is used to store computer-executed instructions; The processor is configured to execute the computer execution instructions to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed, are used to implement the method as described in any one of claims 1 to 7.

Images