Leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring

Abstract

The invention discloses a leakage current monitoring and early warning system based on current monitoring of a new energy photovoltaic inverter, and belongs to the technical field of leakage current monitoring. The invention discloses a leakage current monitoring and early warning system based on current monitoring of a new energy photovoltaic inverter, and the system comprises a sensing module, a data collection and preprocessing module, a central processing module, a multi-stage dynamic threshold module, a complex mode recognition module, and a multi-stage early warning module. And real-time early warning cannot be realized, and a complex leakage current mode cannot be identified. According to the environment parameters monitored in real time and the working state of the inverter, the threshold value of the leakage current is dynamically adjusted, the complex leakage current mode is recognized by adopting a machine learning and mode recognition algorithm in combination with the comprehensive analysis result of the zero-sequence current, the phase and the frequency parameters, and the reliability of the inverter is improved. The multi-level early warning module can give alarms of different levels according to the degree of the leakage current.

Description

Technical Field

[0001] The present invention relates to the technical field of leakage current monitoring, and in particular to a leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring. Background Art

[0002] With the rapid development of new energy technologies, especially the widespread application of photovoltaic power generation systems, the safety of power systems has become a focus of attention. At present, most leakage current monitoring systems use a fixed single threshold setting method. Although this method is simple and easy to use, it has many shortcomings in practical applications. First of all, due to the complex and changeable working environment of the photovoltaic system, a single fixed threshold is difficult to adapt to various working conditions, which can easily lead to false alarms or missed alarms. For example, in a high temperature or high humidity environment, a fixed threshold may cause the system to frequently falsely alarm, while in a low temperature or low humidity environment, the fixed threshold may not be able to detect leakage current in time, thereby increasing safety hazards.

[0003] Secondly, the existing leakage current monitoring system usually lacks real-time warning function. When the system detects leakage current, it can only issue an alarm after the fact, but cannot issue an early warning when the leakage current occurs, which greatly affects the system's response speed and emergency handling capabilities.

[0004] In addition, existing leakage current monitoring technology cannot effectively identify complex leakage current patterns. Many leakage currents are not just simple current anomalies, but are accompanied by changes in multiple factors such as phase, frequency and harmonics. Traditional monitoring systems can often only detect simple current exceeding the standard, but cannot fully analyze these complex changes. Summary of the invention

[0005] The purpose of the present invention is to provide a leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring. By introducing a multi-level dynamic threshold module, according to the real-time monitored environmental parameters and the working state of the inverter, machine learning and pattern recognition algorithms are adopted, combined with the comprehensive analysis results of zero-sequence current, phase and frequency parameters, to identify complex leakage current patterns. Through real-time monitoring and dynamic adjustment of thresholds, the system can timely detect and process leakage currents. The application of multi-sensor fusion and complex pattern recognition algorithms improves the monitoring accuracy and fault diagnosis capability of the system. Through the design of multi-level dynamic threshold adjustment and multi-level early warning mechanism, the false alarm rate is effectively reduced. The threshold is dynamically adjusted according to environmental parameters and working status to adapt to different operating conditions, thereby solving the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring includes:

[0008] The sensor module is installed around the photovoltaic inverter to monitor the operating indicators of the photovoltaic inverter;

[0009] The data acquisition and preprocessing module is connected to the sensor module and is used to perform preliminary filtering and denoising on the sensor data;

[0010] A central processing module, connected to the data acquisition and preprocessing modules, uses a high-performance processor to perform real-time analysis and processing on the preprocessed data;

[0011] A multi-level dynamic threshold module, connected to the central processing module, for real-time monitoring and analysis of environmental parameters and working conditions, and dynamic adjustment of the threshold of the leakage current;

[0012] A complex pattern recognition module, connected to the central processing module, is used to identify complex leakage current patterns by using machine learning and pattern recognition algorithms combined with comprehensive analysis results of zero-sequence current, phase and frequency parameters;

[0013] The multi-level warning module is connected to the central processing module and is used to issue different levels of alarms according to the degree of leakage current after determining that there is leakage current.

[0014] Preferably, the sensor module comprises:

[0015] Temperature sensor, used to monitor ambient temperature in real time;

[0016] Humidity sensor, used to monitor environmental humidity in real time;

[0017] Voltage sensor, used to monitor the changes of grid voltage in real time;

[0018] Current sensor, used to monitor the load size of the inverter in real time;

[0019] Status sensor, used to monitor the working mode of the inverter in real time.

[0020] Preferably, the central processing module is also used for:

[0021] Monitor and calculate zero-sequence current, phase and frequency parameters to determine leakage current mode, fault type and fault location;

[0022] Among them, the zero-sequence current is continuously monitored to determine whether leakage current occurs. When the zero-sequence current exceeds the preset threshold, it is determined that leakage current exists; the phase difference of each phase current is monitored to determine whether the three-phase current is balanced. An unbalanced phase difference indicates the presence of leakage current; the frequency fluctuation is monitored to determine the stability of the power grid. Excessive frequency fluctuation indicates that there is a problem with the power grid;

[0023] Calculate the amplitude and harmonic content of current for load monitoring, system stability assessment, fault diagnosis, energy efficiency evaluation, and model building.

[0024] Preferably, the multi-level dynamic threshold module is further used for:

[0025] Dynamically calculate and adjust the leakage current threshold according to environmental parameters and working conditions;

[0026] The environmental parameters include at least temperature, humidity and grid voltage, and the working state of the inverter includes at least load size and working mode;

[0027] The threshold values ​​under different working conditions are stored in a database;

[0028] Regularly compare the current threshold with historical data to verify the rationality of the current threshold and summarize the long-term trend of threshold changes;

[0029] Provides manual verification function, allowing users to manually adjust the threshold.

[0030] Preferably, the threshold of the leakage current is dynamically calculated and adjusted according to the environmental parameters and working conditions, specifically:

[0031] The threshold is increased in a high temperature environment and decreased in a low temperature environment;

[0032] The threshold is lowered in high humidity environments and increased in low temperature environments;

[0033] Increase the threshold when the grid voltage is high;

[0034] Increase the threshold when the load is high;

[0035] Lowers the threshold in standby mode.

[0036] Preferably, the threshold of the leakage current is dynamically calculated and adjusted according to the environmental parameters and working conditions, specifically:

[0037] The current threshold is calculated and adjusted through linear adjustment, nonlinear adjustment and fuzzy logic algorithm:

[0038] Among them, linear adjustment is adopted for single temperature change, voltage change and load change;

[0039] Adopt nonlinear adjustment for single humidity change and working mode change;

[0040] Fuzzy logic algorithm is used to make adjustments for changes in two or more complex factors among temperature changes, humidity changes, voltage changes, load changes and working mode changes.

[0041] Preferably, the process of the fuzzy logic algorithm is specifically as follows:

[0042] Convert the exact numerical values ​​of temperature, humidity, voltage, load and working mode input variables into membership values ​​of fuzzy sets;

[0043] Match the membership values ​​of input variables according to the fuzzy rule base to determine the applicable rules;

[0044] The outputs of the matched rules are synthesized to obtain fuzzy output;

[0045] The fuzzy output is converted into a precise threshold value and the threshold value of the leakage current is adjusted.

[0046] Preferably, the complex pattern recognition module is further used for:

[0047] Extract the zero-sequence component of the three-phase current, the phase information of each phase current, the frequency information of the current, the amplitude of the current and the harmonic components;

[0048] Select the neural network (ANN) recognition algorithm, use historical data to train the model, adjust model parameters, optimize model performance, and use test set data to verify the accuracy and generalization ability of the model;

[0049] Use trained models to perform pattern recognition on real-time data.

[0050] Preferably, the complex pattern recognition module is also used for:

[0051] Compare the recognition result with the preset threshold to determine whether there is a complex leakage current pattern;

[0052] The corresponding early warning mechanism is triggered based on the recognition results.

[0053] Preferably, the system further comprises:

[0054] User interface, connected to the central processing module, for displaying system status and historical data;

[0055] The communication module is connected to the central processing module and is used to realize data exchange with external devices.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. The present invention introduces a multi-level dynamic threshold module to dynamically adjust the threshold of leakage current according to the real-time monitored environmental parameters and the working state of the inverter. By dynamically adjusting the threshold, the system can accurately identify leakage current under different working conditions, avoid the problem of false alarm or missed alarm caused by fixed threshold, and improve the adaptability and accuracy of the system.

[0058] 2. The present invention adopts machine learning and pattern recognition algorithms, combined with the comprehensive analysis results of zero-sequence current, phase and frequency parameters, to identify complex leakage current patterns. It can identify and process complex leakage current patterns, improve the accuracy and timeliness of fault diagnosis, and ensure the safe operation of the system.

[0059] 3. The present invention has designed a multi-level warning module, which can issue different levels of alarms according to the degree of leakage current. It can issue an early warning as soon as the leakage current occurs, ensure the real-time and response speed of the system, and promptly remind the operator to take corresponding measures to reduce potential safety risks.

[0060] 4. Through real-time monitoring and dynamic adjustment of thresholds, the system of the present invention can timely detect and handle leakage current, reduce potential safety hazards, and improve the overall safety of the system; the application of multi-sensor fusion and complex pattern recognition algorithms improves the monitoring accuracy and fault diagnosis capabilities of the system, ensuring stable operation of the system under various working conditions.

[0061] 5. The present invention effectively reduces the false alarm rate, avoids unnecessary downtime and maintenance, and improves the operating efficiency of the system through the design of multi-level dynamic threshold adjustment and multi-level early warning mechanism; dynamically adjusts the threshold according to environmental parameters and working status to adapt to different operating conditions, thereby improving the adaptability and flexibility of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a module diagram of the leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring of the present invention;

[0063] Figure 2 is a flow chart of the fuzzy logic algorithm of the present invention; DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] In order to solve the problem that the existing leakage current monitoring technology only has a single threshold, cannot achieve real-time warning, and cannot identify complex leakage current patterns, please refer to Figure 1-2 , this embodiment provides the following technical solutions:

[0066] The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring includes:

[0067] The sensor module is installed around the photovoltaic inverter to monitor the operating indicators of the photovoltaic inverter;

[0068] The data acquisition and preprocessing module is connected to the sensor module and is used to perform preliminary filtering and denoising on the sensor data to ensure data quality;

[0069] A central processing module, connected to the data acquisition and preprocessing modules, uses a high-performance processor to perform real-time analysis and processing on the preprocessed data;

[0070] The multi-level dynamic threshold module is connected to the central processing module to monitor and analyze environmental parameters and working conditions in real time and dynamically adjust the threshold of leakage current. Dynamic adjustment can ensure that the system can accurately identify leakage current under different working conditions, avoid false alarms or missed alarms caused by fixed thresholds, and improve the adaptability and accuracy of the system.

[0071] A complex pattern recognition module, connected to the central processing module, is used to identify complex leakage current patterns through machine learning and pattern recognition algorithms combined with comprehensive analysis results of zero-sequence current, phase and frequency parameters;

[0072] The multi-level early warning module is connected to the central processing module and is used to compare the current parameters calculated in real time with the dynamic threshold value to determine whether there is leakage current. When abnormal current is detected, different levels of alarms are issued according to the degree of leakage current to ensure the real-time performance and response speed of the system.

[0073] The user interface is connected to the central processing module and is used to display system status and historical data, which improves the operability and maintainability of the system, facilitates users to monitor and manage the operating status of the inverter in real time, and enhances the user experience and management efficiency of the system;

[0074] The communication module is connected to the central processing module and is used to realize data exchange with external devices to improve system integration and scalability.

[0075] The sensor module includes:

[0076] Temperature sensor, used to monitor ambient temperature in real time, can detect overheating in time to avoid equipment damage;

[0077] Humidity sensor, used to monitor the ambient humidity in real time, can prevent electrical failures caused by excessive humidity;

[0078] Voltage sensors are used to monitor changes in grid voltage and can detect and handle voltage anomalies in a timely manner;

[0079] Current sensor, used to monitor the load size of the inverter to ensure that the system operates within a safe range and avoid overload;

[0080] The status sensor is used to monitor the working mode of the inverter to ensure that the system can work normally in different modes.

[0081] Multi-sensor fusion can comprehensively reflect the operating status and environmental conditions of the inverter, and provide all-round data information for system analysis and decision-making.

[0082] The central processing module is also used to:

[0083] Monitor and calculate zero-sequence current, phase and frequency parameters to determine leakage current mode, fault type and fault location;

[0084] Among them, the zero-sequence current is continuously monitored to determine whether leakage current occurs. When the zero-sequence current exceeds the preset threshold, it is determined that leakage current exists; the phase difference of each phase current is monitored to determine whether the three-phase current is balanced. An unbalanced phase difference indicates the presence of leakage current; the frequency fluctuation is monitored to determine the stability of the power grid. Excessive frequency fluctuation indicates that there is a problem with the power grid;

[0085] Calculate the amplitude and harmonic content of current for load monitoring, system stability assessment, fault diagnosis, energy efficiency evaluation, and model building.

[0086] By comprehensively analyzing the zero-sequence current, phase, frequency, current amplitude and harmonic components, the leakage current mode and fault type can be identified more accurately, improving the diagnostic capability of the system.

[0087] The multi-level dynamic threshold module is further used to:

[0088] Real-time monitoring of environmental parameters, including at least temperature, humidity and grid voltage;

[0089] Real-time monitoring of the inverter's operating status, including at least load size and operating mode;

[0090] Dynamically calculate and adjust the leakage current threshold according to environmental parameters and working conditions;

[0091] The thresholds under different working conditions are stored in the database to facilitate system management and query;

[0092] Regularly compare current thresholds with historical data to ensure the rationality of the thresholds;

[0093] Provides manual verification function, allowing users to manually adjust the threshold to ensure stable system operation.

[0094] The leakage current threshold is dynamically calculated and adjusted according to environmental parameters and working conditions, specifically:

[0095] In high temperature environment, the threshold is increased to avoid false alarms, and in low temperature environment, the threshold is lowered; in high humidity environment, the threshold is lowered to increase sensitivity, and in low temperature environment, the threshold is increased; when the grid voltage is large, the threshold is increased to avoid false alarms; when the load is large, the threshold is increased to avoid false alarms; in standby mode, the threshold is lowered to increase sensitivity.

[0096] The leakage current threshold is dynamically calculated and adjusted according to environmental parameters and working conditions, specifically:

[0097] Linear regulation for single temperature change, voltage change and load change;

[0098] Adopt nonlinear adjustment for single humidity change and working mode change;

[0099] Fuzzy logic algorithm is used to make adjustments for changes in two or more complex factors among temperature changes, humidity changes, voltage changes, load changes and working mode changes.

[0100] The specific process of the fuzzy logic algorithm is as follows:

[0101] Convert the exact numerical values ​​of temperature, humidity, voltage, load and working mode input variables into membership values ​​of fuzzy sets;

[0102] Match the membership values ​​of input variables according to the fuzzy rule base to determine the applicable rules;

[0103] The outputs of the matched rules are synthesized to obtain fuzzy output;

[0104] The fuzzy output is converted into a precise threshold value and the threshold value of the leakage current is adjusted.

[0105] Assume that the current temperature is 26°C and the humidity is 65%.

[0106] An example is provided here, as follows:

[0107] Data collection: Temperature: 26°C, Humidity: 65%

[0108] Temperature normalization: (26-0) / (50-0)=0.52 (assuming the temperature range is 0-50°C)

[0109] Humidity normalization: (65-0) / (100-0)=0.65

[0110] Normalize the temperature and humidity data to the range of 0-1 for easy subsequent processing;

[0111] Temperature membership:

[0112] Low temperature: 0

[0113] Medium temperature: 0.52

[0114] High temperature: 0.48

[0115] Humidity membership:

[0116] Low humidity: 0

[0117] Medium wet: 0.35

[0118] High humidity: 0.65

[0119] Match the corresponding rules from the defined fuzzy rules:

[0120] Rule 1: If the temperature is high and the humidity is high, the threshold is increased by 25%.

[0121] Rule 2: If the temperature is high and the humidity is low, the threshold is increased by 15%.

[0122] Rule 3: If the temperature is low and the humidity is high, the threshold is reduced by 10%.

[0123] Fuzzy synthesis:

[0124] Rule 1 weight: 0.48*0.65=0.312

[0125] Rule 2 weight: 0.48*0.35=0.168

[0126] Rule 3 weight: 0*0.65=0

[0127] Defuzzification:

[0128] Fuzzy output: (0.312*25%)+(0.168*15%)+(0*-10%)=7.8%+2.52%+0%=10.32%

[0129] Threshold adjustment ratio: 10.32%

[0130] Threshold Adjustment:

[0131] Assume the initial threshold is 30mA

[0132] Adjusted threshold: 30mA*(1+10.32%)=30mA*1.1032≈33.096mA

[0133] Through the above steps, the goal of comprehensively adjusting the leakage current threshold according to temperature and humidity is achieved.

[0134] The complex pattern recognition module is further used to:

[0135] Extract the zero-sequence component of the three-phase current, the phase information of each phase current, the frequency information of the current, the amplitude of the current and the harmonic components;

[0136] Select the neural network (ANN) recognition algorithm, use historical data to train the model, adjust model parameters, optimize model performance, and use test set data to verify the accuracy and generalization ability of the model;

[0137] Use trained models to perform pattern recognition on real-time data;

[0138] Compare the recognition result with the preset threshold to determine whether there is a complex leakage current pattern;

[0139] The corresponding early warning mechanism is triggered according to the recognition results to ensure the real-time and response speed of the system.

[0140] Working principle: The system first collects environmental and operating data in real time through sensor modules installed around the inverter. These data are filtered and denoised by the data acquisition and preprocessing module to ensure the accuracy and consistency of the data. The central processing module uses a high-performance processor to analyze the preprocessed data in real time, calculate key parameters such as zero-sequence current, phase and frequency, and determine the leakage current mode and fault type. The multi-level dynamic threshold module dynamically adjusts the threshold of the leakage current according to the real-time monitored environmental parameters and the working status of the inverter. Through linear adjustment, nonlinear adjustment and fuzzy logic algorithm, the system can flexibly adjust the threshold according to the needs under different working conditions to avoid false alarms or missed alarms.

[0141] The complex pattern recognition module uses machine learning and pattern recognition algorithms, combined with the comprehensive analysis results of zero-sequence current, phase and frequency parameters, to identify complex leakage current patterns. The system compares the recognition results with the preset threshold to determine whether a complex leakage current pattern exists, and triggers the multi-level early warning module based on the recognition results to issue different levels of alarms.

[0142] In addition, the system is equipped with a user interface and communication module to display system status and historical data, and to achieve data exchange with external devices to ensure the real-time performance and response speed of the system.

[0143] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0144] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring, characterized in that: include: The sensor module is installed around the photovoltaic inverter to monitor the operating indicators of the photovoltaic inverter; The data acquisition and preprocessing module is connected to the sensor module and is used to perform preliminary filtering and denoising on the sensor data; A central processing module, connected to the data acquisition and preprocessing modules, uses a high-performance processor to perform real-time analysis and processing on the preprocessed data; A multi-level dynamic threshold module, connected to the central processing module, for real-time monitoring and analysis of environmental parameters and working conditions, and dynamic adjustment of the threshold of the leakage current; A complex pattern recognition module, connected to the central processing module, is used to identify complex leakage current patterns by using machine learning and pattern recognition algorithms combined with comprehensive analysis results of zero-sequence current, phase and frequency parameters; The multi-level warning module is connected to the central processing module and is used to issue different levels of alarms according to the degree of leakage current after determining that there is leakage current.

2. The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring according to claim 1 is characterized in that: The sensor module comprises: Temperature sensor, used to monitor ambient temperature in real time; Humidity sensor, used to monitor environmental humidity in real time; Voltage sensor, used to monitor the changes of grid voltage in real time; Current sensor, used to monitor the load size of the inverter in real time; Status sensor, used to monitor the working mode of the inverter in real time.

3. The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring according to claim 2 is characterized in that: The central processing module is also used for: Monitor and calculate zero-sequence current, phase and frequency parameters to determine leakage current mode, fault type and fault location; Among them, the zero-sequence current is continuously monitored to determine whether leakage current occurs. When the zero-sequence current exceeds the preset threshold, it is determined that leakage current exists; the phase difference of each phase current is monitored to determine whether the three-phase current is balanced. An unbalanced phase difference indicates the presence of leakage current; the frequency fluctuation is monitored to determine the stability of the power grid. Excessive frequency fluctuation indicates that there is a problem with the power grid; Calculate the amplitude and harmonic content of current for load monitoring, system stability assessment, fault diagnosis, energy efficiency evaluation, and model building.

4. The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring according to claim 3 is characterized in that: The multi-level dynamic threshold module is further used for: Dynamically calculate and adjust the leakage current threshold according to environmental parameters and working conditions; The environmental parameters include at least temperature, humidity and grid voltage, and the working state of the inverter includes at least load size and working mode; The threshold values ​​under different working conditions are stored in a database; Regularly compare the current threshold with historical data to verify the rationality of the current threshold and summarize the long-term trend of threshold changes; Provides manual verification function, allowing users to manually adjust the threshold.

5. The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring according to claim 4 is characterized in that: The threshold of leakage current is dynamically calculated and adjusted according to environmental parameters and working conditions, specifically: The threshold is increased in a high temperature environment and decreased in a low temperature environment; The threshold is lowered in high humidity environments and increased in low temperature environments; Increase the threshold when the grid voltage is high; Increase the threshold when the load is high; Lowers the threshold in standby mode.

6. The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring according to claim 5 is characterized in that: The threshold of leakage current is dynamically calculated and adjusted according to environmental parameters and working conditions, specifically: The current threshold is calculated and adjusted through linear adjustment, nonlinear adjustment and fuzzy logic algorithm: Among them, linear adjustment is adopted for single temperature change, voltage change and load change; Adopt nonlinear adjustment for single humidity change and working mode change; Fuzzy logic algorithm is used to make adjustments for changes in two or more complex factors among temperature changes, humidity changes, voltage changes, load changes and working mode changes.

7. The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring according to claim 6 is characterized in that: The process of the fuzzy logic algorithm is specifically as follows: Convert the exact numerical values ​​of temperature, humidity, voltage, load and working mode input variables into membership values ​​of fuzzy sets; Match the membership values ​​of input variables according to the fuzzy rule base to determine the applicable rules; The outputs of the matched rules are synthesized to obtain fuzzy output; The fuzzy output is converted into a precise threshold value and the threshold value of the leakage current is adjusted.

8. The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring according to claim 7 is characterized in that: The complex pattern recognition module is further used for: Extract the zero-sequence component of the three-phase current, the phase information of each phase current, the frequency information of the current, the amplitude of the current and the harmonic components; Select a neural network recognition algorithm, use historical data to train the model, adjust model parameters, optimize model performance, and use test set data to verify the accuracy and generalization ability of the model; Use trained models to perform pattern recognition on real-time data.

9. The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring according to claim 8 is characterized in that: The complex pattern recognition module is also used for: Compare the recognition result with the preset threshold to determine whether there is a complex leakage current pattern; The corresponding early warning mechanism is triggered based on the recognition results.

10. The leakage current monitoring and early warning system based on new energy photovoltaic inverter current monitoring according to claim 9 is characterized in that: The system further comprises: User interface, connected to the central processing module, for displaying system status and historical data; The communication module is connected to the central processing module and is used to realize data exchange with external devices.

Images