A method for analyzing stability anomalies in virtual synchronous generators

Abstract

This invention relates to the field of power grid monitoring technology and discloses a method for analyzing the stability anomalies of a virtual synchronous generator, comprising the following steps: Step 1: Real-time acquisition of operational data of the virtual synchronous generator and the power grid system at all time points, and classification into datasets; Step 2: Setting monitoring cycles and reference thresholds to generate corresponding abnormal data groups; Step 3: Establishing a time axis based on the monitoring cycle, and then inserting corresponding nodes into the abnormal data groups to analyze the variation characteristics of all nodes in the frequency and time domains, and generating corresponding fluctuation indices and distribution frequencies, flexibly adapting to load changes and exhibiting strong stability; Step 4: Setting fluctuation thresholds and distribution thresholds to determine whether the virtual synchronous generator has operational faults, generating corresponding decision trees according to the monitoring cycle, diagnosing the causes of faults during the operation of the virtual synchronous generator in real time, and notifying maintenance and management personnel, with high accuracy in real-time fault diagnosis.

Description

Technical Field

[0001] This invention relates to the field of power grid monitoring technology, specifically to a method for analyzing stability anomalies in virtual synchronous generators. Background Technology

[0002] A virtual synchronous generator (VSG) is a technology that simulates the operating characteristics of a traditional synchronous generator. It uses inverter control to mimic the mechanical and electromagnetic characteristics of a synchronous generator, providing the necessary inertia and damping for the power grid. In large-scale renewable energy grid integration, the VSG simulates the active power frequency regulation and reactive power voltage regulation functions of a synchronous generator, participating in grid frequency and voltage regulation. By absorbing excess renewable energy in the grid, it achieves peak shaving and valley filling, optimizes electricity consumption, improves grid stability and compatibility, and reduces the impact of renewable energy generation fluctuations on the grid. During grid faults, power angle instability can easily occur due to improper system parameters or control strategies. Developing targeted adaptive transient control strategies can reduce active power imbalance and ensure the reliable operation of the VSG in the power system. A deeper understanding of the causes and solutions to VSG stability anomalies will help promote the development of a more efficient and stable power system.

[0003] Currently, traditional methods for analyzing the stability anomalies of virtual synchronous generators lack flexibility and adaptability. Faced with complex load changes in the power grid system, the algorithm model is relatively simple, the data analysis capability is limited, and it cannot detect whether there are operational faults in the virtual synchronous generator in a timely manner, nor can it accurately and reliably locate the cause of the fault. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for analyzing the stability anomalies of virtual synchronous generators. This method offers advantages such as strong stability in adapting to load changes and high accuracy in real-time fault diagnosis, thus solving the problems of traditional methods for analyzing the stability anomalies of virtual synchronous generators, which lack flexibility and are not accurate or reliable enough in locating the causes of faults.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for analyzing stability anomalies in a virtual synchronous generator, comprising the following steps:

[0006] Step 1: Real-time operation data of the virtual synchronous generator and the power grid system at all points in time are acquired through network connection sensing devices. The acquired data is then classified into power datasets and load datasets. The power dataset includes the real-time speed, real-time current, real-time voltage, and real-time frequency output by the virtual synchronous generator. The load dataset includes the real-time current, real-time voltage, and real-time frequency required by the power grid system.

[0007] Step 2: Set a fixed monitoring period Then, regarding the monitoring cycle Set a reference threshold within a fixed range. Based on the power dataset and load dataset, identify the monitoring period. The abnormal operation data within the data will generate corresponding abnormal data groups. ;

[0008] Step 3: According to the monitoring cycle Establish a timeline Combined with abnormal data groups Insert the corresponding nodes, analyze the variation characteristics of all nodes in the frequency and time domains, and generate the corresponding fluctuation index. and distribution frequency ;

[0009] Step 4: Set a fixed range of fluctuation thresholds and distribution threshold And combined with volatility index and distribution frequency To determine whether the virtual synchronous generator has an operational fault, according to the monitoring cycle. Divide the dataset into training and test sets, and generate corresponding decision trees. It can diagnose the causes of faults in the virtual synchronous generator during operation in real time and notify maintenance and management personnel.

[0010] Preferably, in step one, the expression for the power dataset is: , This represents the real-time rotational speed output by the virtual synchronous generator. This represents the real-time output current of the virtual synchronous generator. This represents the real-time voltage output by the virtual synchronous generator. This represents the real-time frequency output by the virtual synchronous generator. This indicates the time point at which the virtual synchronous generator's operating data was acquired.

[0011] Preferably, in step one, the expression for the load dataset is: , This represents the real-time current required by the power grid system. This indicates the real-time voltage required by the power grid system. This indicates the real-time frequency required by the power grid system. This indicates the time point at which the power grid system operation data was acquired.

[0012] Preferably, in step two, the reference threshold Including speed threshold Current threshold Voltage threshold and frequency threshold .

[0013] Preferably, in step two, the abnormal data group The calculation process is as follows:

[0014] Based on the power dataset, the monitoring period is extracted. Real-time speed output of the internal virtual synchronous generator Real-time current Real-time voltage and real-time frequency ;

[0015] Based on the load dataset, extract the monitoring period. Real-time current required by the internal power grid system Real-time voltage and real-time frequency ;

[0016] ;

[0017] In the formula, Indicates abnormal data groups. This represents the minimum speed threshold. This indicates the maximum speed threshold, if the monitoring period is... Real-time speed output of the internal virtual synchronous generator When the speed exceeds the threshold, it is marked as an abnormal speed. This represents the minimum current threshold value. This indicates the maximum current threshold value, if the monitoring period is... Real-time current output of the internal virtual synchronous generator When the current exceeds the threshold, it is marked as an abnormal output current. This represents the minimum voltage threshold value. This indicates the maximum voltage threshold value, if the monitoring period is... Real-time voltage output of the internal virtual synchronous generator When the voltage exceeds the threshold, it is marked as an abnormal output voltage. This represents the minimum frequency threshold value. This indicates the maximum frequency threshold value, if the monitoring period is... Real-time frequency output of the internal virtual synchronous generator When the frequency exceeds the threshold, it is marked as an abnormal output frequency. (If the monitoring period...) Real-time current required by the internal power grid system When the current exceeds the threshold, it is marked as an abnormal load current. (If the monitoring period...) Real-time voltage required by the internal power grid system When the voltage exceeds the threshold, it is marked as an abnormal load voltage. (If the monitoring period...) Real-time frequency required by internal power grid system When the frequency exceeds the threshold, it is marked as an abnormal load frequency.

[0018] Preferably, in step three, the volatility index The calculation process is as follows:

[0019] abnormal data group The abnormal speed, abnormal output current, abnormal output voltage, abnormal output frequency, abnormal load current, abnormal load voltage, or abnormal load frequency marked in the middle are inserted into the time axis. And generate one-to-one corresponding nodes;

[0020] If abnormal data group Marked An abnormal output current, The expression for the abnormal output current is: , to The time axis represents each abnormal output current in sequence. The corresponding node in;

[0021] ;

[0022] ;

[0023] In the formula, Indicates volatility index, The sum of abnormal output currents divided by the number of nodes yields the average value of the abnormal output currents. Representing the timeline The Middle The abnormal output current corresponding to each node. This indicates that, according to the variance formula, it is calculated simultaneously. The variance obtained by taking the difference between each abnormal output current and its average value is the fluctuation index of the abnormal output current in the frequency domain.

[0024] Preferably, in step three, the distribution frequency The calculation process is as follows:

[0025] According to the timeline Calculate the time difference between every two adjacent abnormal output currents and mark it as... , to In order to represent The time difference between two adjacent abnormal output currents in each node;

[0026] ;

[0027] In the formula, Indicates the distribution frequency. express The sum of the time differences of the groups This indicates the frequency of occurrence of abnormal output current in the time domain, and also indicates... The probability distribution of abnormal output current nodes in the time domain.

[0028] Preferably, in step four, the volatility index is... Comparison of fluctuation thresholds If a single monitoring cycle Internal volatility index Exceeding the fluctuation threshold When this occurs, it indicates that the virtual synchronous generator has an operational fault.

[0029] Preferably, in step four, the distribution frequency is... Comparison distribution threshold If a single monitoring cycle Internal distribution frequency Exceeding the distribution threshold When this occurs, it indicates that the virtual synchronous generator has an operational fault.

[0030] Preferably, in step four, the decision tree... The training process is as follows:

[0031] Divided chronologically from earliest to latest Each monitoring cycle The operating data of the internal virtual synchronous generator and the power grid system will be used to... Each monitoring cycle The running data was used as the training set, and then... Each monitoring cycle The runtime data is used as the test set, with reference thresholds. Fluctuation threshold and distribution threshold As a decision tree hyperparameters;

[0032] Decision trees are constructed using the training set. The main framework is then used to generate corresponding decision trees based on the data types in the power and load datasets. Branches, and each branch corresponds to a set of abnormal data groups. Volatility Index Distribution frequency And the results of the operational fault diagnosis;

[0033] Evaluate the decision tree using the test set. The accuracy rate of generating operational fault diagnosis results; if the accuracy rate reaches... , indicating that training generates decision trees It has real-time diagnostic capabilities; if the accuracy rate does not reach [a certain level], [it will not be able to achieve the desired , indicating that training generates decision trees Insufficient real-time diagnostic capabilities necessitate the addition of multiple monitoring cycles. The running data is used to continuously update the decision tree. Until the accuracy rate reaches .

[0034] Compared with existing technologies, this invention provides a method for analyzing stability anomalies in virtual synchronous generators, which has the following beneficial effects:

[0035] 1. This invention acquires real-time operational data of the virtual synchronous generator and the power grid system at all points in time through network-connected sensing devices. The acquired data is then categorized into power datasets and load datasets, facilitating rapid assessment of the supply-demand balance between the virtual synchronous generator and the power grid system, and allowing for the setting of fixed-duration monitoring periods. Then, regarding the monitoring cycle Set a reference threshold within a fixed range. Based on the power dataset and load dataset, identify the monitoring period. The abnormal operation data within the data will generate corresponding abnormal data groups. According to the monitoring cycle Establish a timeline Combined with abnormal data groups Insert the corresponding nodes, analyze the variation characteristics of all nodes in the frequency and time domains, and calculate the fluctuation index in the frequency and time domains in sequence according to the types of outlier values. and distribution frequency It is flexible and adaptable to load changes with strong stability.

[0036] 2. This invention sets a fluctuation threshold within a fixed range. and distribution threshold Volatility Index Comparison of fluctuation thresholds If a single monitoring cycle Internal volatility index Exceeding the fluctuation threshold When this occurs, it indicates that the virtual synchronous generator has an operational fault, and the distributed frequency will be... Comparison distribution threshold If a single monitoring cycle Internal distribution frequency Exceeding the distribution threshold When this occurs, it indicates that the virtual synchronous generator has an operational fault, according to the monitoring cycle. Divide the dataset into training and test sets, and generate corresponding decision trees. The decision tree obtained through training The accuracy rate of judgment must reach Decision tree Each branch corresponds to a set of abnormal data groups Volatility Index Distribution frequency The results of fault diagnosis and analysis ensure the effectiveness of the decision tree. The accuracy of each branch is high, enabling real-time diagnosis of fault causes during the operation of the virtual synchronous generator and notification to maintenance personnel. Attached Figure Description

[0037] Figure 1 This is a diagram illustrating the steps of the method of the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Traditional methods for analyzing stability anomalies in virtual synchronous generators lack flexibility and adaptability. Faced with complex load changes in power grid systems, their algorithm models are relatively simple, and their data analysis capabilities are limited. They cannot promptly detect operational faults in virtual synchronous generators, and their fault location is not accurate or reliable enough. Therefore, this paper proposes a new method for analyzing stability anomalies in virtual synchronous generators. Please refer to [link to relevant documentation]. Figure 1 The method includes the following steps:

[0040] Step 1: Real-time operational data of the virtual synchronous generator and the power grid system at all points in time is acquired via network-connected sensing devices. The acquired data is then categorized into a power dataset and a load dataset. The power dataset includes the real-time speed, current, voltage, and frequency output by the virtual synchronous generator. The expression for the power dataset is as follows: , This represents the real-time rotational speed output by the virtual synchronous generator. This represents the real-time output current of the virtual synchronous generator. This represents the real-time voltage output by the virtual synchronous generator. This represents the real-time frequency output by the virtual synchronous generator. This indicates the time point at which the virtual synchronous generator's operating data is acquired. The load dataset includes the real-time current, real-time voltage, and real-time frequency required by the power grid system. The expression for the load dataset is: , This represents the real-time current required by the power grid system. This indicates the real-time voltage required by the power grid system. This indicates the real-time frequency required by the power grid system. This indicates the time point at which the power grid system operation data is acquired. Comprehensive collection of real-time operation data of the virtual synchronous generator and the power grid system is beneficial for quickly determining whether the supply and demand relationship between the virtual synchronous generator and the power grid system is balanced.

[0041] Step 2: Set a fixed monitoring period Then, regarding the monitoring cycle Set a reference threshold within a fixed range. The reference thresholds include the speed threshold. Current threshold Voltage threshold and frequency threshold Based on the power dataset and load dataset, identify the monitoring period. The abnormal operation data within the data will generate corresponding abnormal data groups. The calculation process is as follows:

[0042] Based on the power dataset, the monitoring period is extracted. Real-time speed output of the internal virtual synchronous generator Real-time current Real-time voltage and real-time frequency ;

[0043] Based on the load dataset, extract the monitoring period. Real-time current required by the internal power grid system Real-time voltage and real-time frequency ;

[0044] ;

[0045] In the formula, Indicates abnormal data groups. This represents the minimum speed threshold. This indicates the maximum speed threshold, if the monitoring period is... Real-time speed output of the internal virtual synchronous generator When the speed exceeds the threshold, it is marked as an abnormal speed. This represents the minimum current threshold value. This indicates the maximum current threshold value, if the monitoring period is... Real-time current output of the internal virtual synchronous generator When the current exceeds the threshold, it is marked as an abnormal output current. This represents the minimum voltage threshold value. This indicates the maximum voltage threshold value, if the monitoring period is... Real-time voltage output of the internal virtual synchronous generator When the voltage exceeds the threshold, it is marked as an abnormal output voltage. This represents the minimum frequency threshold value. This indicates the maximum frequency threshold value, if the monitoring period is... Real-time frequency output of the internal virtual synchronous generator When the frequency exceeds the threshold, it is marked as an abnormal output frequency. (If the monitoring period...) Real-time current required by the internal power grid system When the current exceeds the threshold, it is marked as an abnormal load current. (If the monitoring period...) Real-time voltage required by the internal power grid system When the voltage exceeds the threshold, it is marked as an abnormal load voltage. (If the monitoring period...) Real-time frequency required by internal power grid system When the frequency threshold is exceeded, it is marked as an abnormal load frequency. A preliminary analysis of the real-time operating data is performed to quickly identify abnormal data, which facilitates further in-depth analysis of the abnormal characteristics of the data.

[0046] Step 3: According to the monitoring cycle Establish a timeline Combined with abnormal data groups Insert the corresponding nodes, analyze the variation characteristics of all nodes in the frequency and time domains, and generate the corresponding fluctuation index. and distribution frequency The calculation process is as follows:

[0047] abnormal data group The abnormal speed, abnormal output current, abnormal output voltage, abnormal output frequency, abnormal load current, abnormal load voltage, or abnormal load frequency marked in the middle are inserted into the time axis. And generate one-to-one corresponding nodes. In actual use, each abnormal value corresponds to a node of a certain color. Subsequent analysis will be conducted based on the node data of the same color.

[0048] If abnormal data group Marked An abnormal output current, The expression for the abnormal output current is: , to The time axis represents each abnormal output current in sequence. The corresponding node in;

[0049] ;

[0050] In the formula, Indicates volatility index, The sum of abnormal output currents divided by the number of nodes yields the average value of the abnormal output currents. Representing the timeline The Middle The abnormal output current corresponding to each node. This indicates that, according to the variance formula, it is calculated simultaneously. The variance obtained by calculating the difference between each abnormal output current and its average value is the fluctuation index of the abnormal output current in the frequency domain. Similarly, the fluctuation index in the frequency domain is calculated for each abnormal value using the above formula. ;

[0051] According to the timeline Calculate the time difference between every two adjacent abnormal output currents and mark it as... , to In order to represent The time difference between two adjacent abnormal output currents in each node;

[0052] ;

[0053] In the formula, Indicates the distribution frequency. express The sum of the time differences of the groups This indicates the frequency of occurrence of abnormal output current in the time domain, and also indicates... The probability distribution of each abnormal output current node in the time domain is calculated using the above formula, and so on. The distribution frequency in the time domain for each abnormal value is calculated according to the above formula. It is flexible and adaptable to load changes with strong stability;

[0054] Step 4: Set a fixed range of fluctuation thresholds and distribution threshold , will volatility index Comparison of fluctuation thresholds If a single monitoring cycle Internal volatility index Exceeding the fluctuation threshold When this occurs, it indicates that the virtual synchronous generator has an operational fault, and the distributed frequency will be... Comparison distribution threshold If a single monitoring cycle Internal distribution frequency Exceeding the distribution threshold When this occurs, it indicates that the virtual synchronous generator has an operational fault;

[0055] According to the monitoring cycle Divide the dataset into training and test sets, and generate corresponding decision trees. The training process is as follows:

[0056] Divided chronologically from earliest to latest Each monitoring cycle The operating data of the internal virtual synchronous generator and the power grid system will be used to... Each monitoring cycle The running data was used as the training set, and then... Each monitoring cycle The runtime data is used as the test set, with reference thresholds. Fluctuation threshold and distribution threshold As a decision tree hyperparameters;

[0057] Decision trees are constructed using the training set. The main framework is then used to generate corresponding decision trees based on the data types in the power and load datasets. The virtual synchronous generator outputs real-time speed, real-time current, real-time voltage, and real-time frequency, as well as the real-time current, real-time voltage, and real-time frequency required by the power grid system, corresponding to six branches. Each branch also corresponds to a set of abnormal data. Volatility Index Distribution frequency And the results of the operational fault diagnosis;

[0058] Evaluate the decision tree using the test set. The accuracy rate of generating operational fault diagnosis results; if the accuracy rate reaches... , indicating that training generates decision trees It has real-time diagnostic capabilities; if the accuracy rate does not reach [a certain level], [it will not be able to achieve the desired , indicating that training generates decision trees Insufficient real-time diagnostic capabilities necessitate the addition of multiple monitoring cycles. The running data is used to continuously update the decision tree. Until the accuracy rate reaches This ensures the decision tree The accuracy of each branch is high, enabling real-time diagnosis of fault causes during the operation of the virtual synchronous generator and notification to maintenance personnel.

[0059] Example 1: In this experiment, 10 hours of operating data from a virtual synchronous generator and a power grid system were selected as the experimental subjects. Statistical analysis revealed that the reference thresholds were: speed 3000-4000 RPM, current 1500-2000 A, voltage 210-230 V, and frequency 50-60 Hz. Five abnormal output current values ​​(1200 A, 1280 A, 1440 A, 1390 A, and 1300 A) were marked in the 10 hours of virtual synchronous generator operating data. Each abnormal output current value has a corresponding node on the time axis. The formula for calculating the fluctuation index corresponding to the abnormal output current values ​​of the virtual synchronous generator within 10 hours is as follows:

[0060] ;

[0061] In the formula, Indicates volatility index, This represents the average value of the abnormal output current. This represents the variance value obtained by simultaneously calculating the difference between the five abnormal output current values ​​and the average value according to the variance formula. This is the fluctuation index of abnormal output current in the frequency domain.

[0062] Example 2: In this experiment, 6 hours of operating data from a virtual synchronous generator and the power grid system were used as the experimental subjects. Statistical analysis revealed four abnormal output current values ​​(1800A, 2000A, 1500A, and 1930A) in the 6 hours of virtual synchronous generator operating data. Each abnormal output current value had a corresponding node on the time axis. The time difference between any two adjacent abnormal output currents was 37 minutes, 66 minutes, and 10 minutes. The distribution frequency of the abnormal output current values ​​of the virtual synchronous generator within 6 hours was also analyzed. The calculation formula is as follows:

[0063] ;

[0064] In the formula, The distribution frequency is represented by 38 min / time, which indicates the frequency of occurrence of abnormal output current in the time domain, and also represents the distribution probability of these four abnormal output current nodes in the time domain.

[0065] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for analyzing stability anomalies in a virtual synchronous generator, characterized in that: Includes the following steps: Step 1: Real-time operation data of the virtual synchronous generator and the power grid system at all points in time are acquired through network connection sensing devices. The acquired data is then classified into power datasets and load datasets. The power dataset includes the real-time speed, real-time current, real-time voltage, and real-time frequency output by the virtual synchronous generator. The load dataset includes the real-time current, real-time voltage, and real-time frequency required by the power grid system. Step 2: Set a fixed monitoring period Then, regarding the monitoring cycle Set a reference threshold within a fixed range. Based on the power dataset and load dataset, identify the monitoring period. The abnormal operation data within the data will generate corresponding abnormal data groups. ; Step 3: According to the monitoring cycle Establish a timeline Combined with abnormal data groups Insert the corresponding nodes, analyze the variation characteristics of all nodes in the frequency and time domains, and generate the corresponding fluctuation index. and distribution frequency Distribution frequency The calculation process is as follows: According to the timeline Calculate the time difference between every two adjacent abnormal output currents and mark it as... , to In order to represent The time difference between two adjacent abnormal output currents in each node; ； In the formula, Indicates the distribution frequency. express The sum of the time differences of the groups This indicates the frequency of occurrence of abnormal output current in the time domain, and also indicates... The probability distribution of each abnormal output current node in the time domain; Step 4: Set a fixed range of fluctuation thresholds and distribution threshold And combined with volatility index and distribution frequency To determine whether the virtual synchronous generator has an operational fault, according to the monitoring cycle. Divide the dataset into training and test sets, and generate corresponding decision trees. It can diagnose the causes of faults in the virtual synchronous generator during operation in real time and notify maintenance and management personnel.

2. The method for analyzing stability anomalies of a virtual synchronous generator according to claim 1, characterized in that: In step one, the expression for the power dataset is: , This represents the real-time rotational speed output by the virtual synchronous generator. This represents the real-time output current of the virtual synchronous generator. This represents the real-time voltage output by the virtual synchronous generator. This represents the real-time frequency output by the virtual synchronous generator. This indicates the time point at which the virtual synchronous generator's operating data was acquired.

3. The method for analyzing stability anomalies of a virtual synchronous generator according to claim 2, characterized in that: In step one, the expression for the load dataset is: , This represents the real-time current required by the power grid system. This indicates the real-time voltage required by the power grid system. This indicates the real-time frequency required by the power grid system. This indicates the time point at which the power grid system operation data was acquired.

4. The method for analyzing stability anomalies of a virtual synchronous generator according to claim 3, characterized in that: In step two, the reference threshold Including speed threshold Current threshold Voltage threshold and frequency threshold .

5. The method for analyzing stability anomalies of a virtual synchronous generator according to claim 4, characterized in that: In step two, the abnormal data group The calculation process is as follows: Based on the power dataset, the monitoring period is extracted. Real-time speed output of the internal virtual synchronous generator Real-time current Real-time voltage and real-time frequency ; Based on the load dataset, extract the monitoring period. Real-time current required by the internal power grid system Real-time voltage and real-time frequency ; ； In the formula, Indicates abnormal data groups. This represents the minimum speed threshold. This indicates the maximum speed threshold, if the monitoring period is... Real-time speed output of the internal virtual synchronous generator When the speed exceeds the threshold, it is marked as an abnormal speed. This represents the minimum current threshold value. This indicates the maximum current threshold value, if the monitoring period is... Real-time current output of the internal virtual synchronous generator When the current exceeds the threshold, it is marked as an abnormal output current. This represents the minimum voltage threshold value. This indicates the maximum voltage threshold value, if the monitoring period is... Real-time voltage output of the internal virtual synchronous generator When the voltage exceeds the threshold, it is marked as an abnormal output voltage. This represents the minimum frequency threshold value. This indicates the maximum frequency threshold value, if the monitoring period is... Real-time frequency output of the internal virtual synchronous generator When the frequency exceeds the threshold, it is marked as an abnormal output frequency. (If the monitoring period...) Real-time current required by the internal power grid system When the current exceeds the threshold, it is marked as an abnormal load current. (If the monitoring period...) Real-time voltage required by the internal power grid system When the voltage exceeds the threshold, it is marked as an abnormal load voltage. (If the monitoring period...) Real-time frequency required by internal power grid system When the frequency exceeds the threshold, it is marked as an abnormal load frequency.

6. The method for analyzing stability anomalies of a virtual synchronous generator according to claim 5, characterized in that: In step three, the volatility index The calculation process is as follows: abnormal data group The abnormal speed, abnormal output current, abnormal output voltage, abnormal output frequency, abnormal load current, abnormal load voltage, or abnormal load frequency marked in the middle are inserted into the time axis. And generate one-to-one corresponding nodes; If abnormal data group Marked An abnormal output current, The expression for the abnormal output current is: , to The time axis represents each abnormal output current in sequence. The corresponding node in; ; ； In the formula, Indicates volatility index, The sum of abnormal output currents divided by the number of nodes yields the average value of the abnormal output currents. Representing the timeline The Middle The abnormal output current corresponding to each node. This indicates that, according to the variance formula, it is calculated simultaneously. The variance obtained by taking the difference between each abnormal output current and its average value is the fluctuation index of the abnormal output current in the frequency domain.

7. The method for analyzing stability anomalies of a virtual synchronous generator according to claim 1, characterized in that: In step four, the volatility index Comparison of fluctuation thresholds If a single monitoring cycle Internal volatility index Exceeding the fluctuation threshold When this occurs, it indicates that the virtual synchronous generator has an operational fault.

8. The method for analyzing stability anomalies of a virtual synchronous generator according to claim 7, characterized in that: In step four, the distribution frequency Comparison distribution threshold If a single monitoring cycle Internal distribution frequency Exceeding the distribution threshold When this occurs, it indicates that the virtual synchronous generator has an operational fault.

9. The method for analyzing stability anomalies of a virtual synchronous generator according to claim 8, characterized in that: In step four, the decision tree The training process is as follows: Divided chronologically from earliest to latest Each monitoring cycle The operating data of the internal virtual synchronous generator and the power grid system will be used to... Each monitoring cycle The running data was used as the training set, and then... Each monitoring cycle The runtime data is used as the test set, with reference thresholds. Fluctuation threshold and distribution threshold As a decision tree hyperparameters; Decision trees are constructed using the training set. The main framework is then used to generate corresponding decision trees based on the data types in the power and load datasets. Branches, and each branch corresponds to a set of abnormal data groups. Volatility Index Distribution frequency And the results of the operational fault diagnosis; Evaluate the decision tree using the test set. The accuracy rate of generating operational fault diagnosis results; if the accuracy rate reaches... , indicating that training generates decision trees It has real-time diagnostic capabilities; if the accuracy rate does not reach [a certain level], [it will not be able to achieve the desired result]. , indicating that training generates decision trees Insufficient real-time diagnostic capabilities necessitate the addition of multiple monitoring cycles. The running data is used to continuously update the decision tree. Until the accuracy rate reaches .

Images