Circuit protection method and system based on real-time current data

By real-time acquisition and preprocessing of the current data of the subway DC power supply system, extracting characteristic parameters for fault diagnosis and dynamic protection threshold adjustment, and formulating multi-level protection strategies, it solves the difficulty of protection methods in the existing technology to adapt to dynamic load changes and complex faults. The problem of this problem improves the timeliness and accuracy of protection and ensures system security.

CN118889318BActive Publication Date: 2024-12-06CCCC (TIANJIN) TECH INSPECTION CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411375894.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-30
Publication Date
2024-12-06
Estimated Expiration
2044-09-30

AI Technical Summary

Technical Problem

The existing protection methods of subway DC power supply systems are difficult to adapt to dynamic load changes and complex fault conditions. Especially in the face of harmonic pollution and rapid transients introduced by new power electronic equipment, the protection effect is not ideal, and real-time data analysis is lacking and Adaptive adjustment capability.

Method used

By real-time acquisition and preprocessing of the current data of the subway DC power supply system, extracting characteristic parameters, troubleshooting and dynamic protection threshold adjustment, multi-level protection strategies are formulated, including early warning, current limiting and disconnection protection, and protection is updated according to real-time status. Strategy.

Benefits of technology

Improve the timeliness and accuracy of protection, enhance the ability to identify faults of varying degrees, avoid excessive or insufficient protection problems caused by fixed thresholds, ensure system safety and reduce unnecessary power outages.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118889318B_ABST
    Figure CN118889318B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of data processing technology, and discloses a circuit protection method and system based on real-time current data, which are used to improve the efficiency and accuracy of circuit protection based on real-time current data. The method includes: extracting and analyzing the features of the pre-processed current data to obtain current feature parameters; performing fault diagnosis on the current feature parameters to obtain a fault type judgment result, which includes normal operation, slight abnormality, and severe fault; performing comprehensive analysis on the fault type judgment result and the current operation state to obtain a dynamic protection threshold; analyzing the dynamic protection threshold and the current feature parameters to obtain a DC power supply system protection strategy, which includes early warning, current limiting, and disconnection protection; performing protection control on the subway DC power supply system according to the DC power supply system protection strategy to obtain a protection execution result, and analyzing the protection execution result to update the DC power supply system protection strategy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a circuit protection method and system based on real-time current data. Background Art

[0002] The subway DC power supply system is a key infrastructure of urban rail transit, and its safety and reliability directly affect the operation of the entire subway system. Traditional DC power supply system protection methods mainly rely on fixed threshold overcurrent protection, differential protection and distance protection. These methods judge the fault state through preset protection values ​​and trigger corresponding protection actions. At the same time, some advanced protection schemes have introduced microcomputer protection devices and used digital signal processing technology to improve the accuracy and response speed of protection. In addition, some subway systems have also adopted communication-based coordinated protection strategies to achieve more flexible protection coordination through information interaction between devices.

[0003] However, the existing protection methods still have some shortcomings. First, the fixed threshold protection strategy is difficult to adapt to the dynamic changes of subway load and different operating conditions, which can easily lead to insufficient protection sensitivity or false operation. Secondly, although the microcomputer protection device has improved the data processing capability, the judgment logic in complex fault conditions is still insufficient, especially for transient processes and multiple faults. The recognition ability is limited. Furthermore, the existing coordinated protection strategy is mainly based on preset logic and lacks the ability to deeply analyze and adaptively adjust real-time operating data. Finally, traditional protection methods often show less than ideal protection effects when faced with problems such as harmonic pollution and fast transients after the introduction of new power electronic equipment. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a circuit protection method and system based on real-time current data, which are used to improve the efficiency and accuracy of circuit protection based on real-time current data.

[0005] The present invention provides a circuit protection method based on real-time current data, comprising: real-time acquisition and preprocessing of current data of a subway direct current power supply system to obtain preprocessed current data; feature extraction and analysis of the preprocessed current data to obtain current characteristic parameters; fault diagnosis of the current characteristic parameters to obtain a fault type judgment result, wherein the fault type judgment result includes normal operation, slight abnormality and serious fault; comprehensive analysis of the fault type judgment result and the current operation state to obtain a dynamic protection threshold; analysis of the dynamic protection threshold and the current characteristic parameters to obtain a direct current power supply system protection strategy, wherein the direct current power supply system protection strategy includes early warning, current limiting and disconnection protection; protection control of the subway direct current power supply system according to the direct current power supply system protection strategy to obtain a protection execution result, and analysis of the protection execution result to update the direct current power supply system protection strategy.

[0006] The present invention also provides a circuit protection system based on real-time current data, comprising:

[0007] A processing module is used to collect and preprocess the current data of the subway DC power supply system in real time to obtain the preprocessed current data;

[0008] An extraction module, used for performing feature extraction and analysis on the preprocessed current data to obtain current characteristic parameters;

[0009] A diagnosis module, used to perform fault diagnosis on the current characteristic parameters to obtain a fault type judgment result, wherein the fault type judgment result includes normal operation, slight abnormality and serious fault;

[0010] An analysis module is used to comprehensively analyze the fault type judgment result and the current operating state to obtain a dynamic protection threshold;

[0011] A protection module, used for analyzing the dynamic protection threshold and the current characteristic parameter to obtain a DC power supply system protection strategy, wherein the DC power supply system protection strategy includes early warning, current limiting and disconnection protection;

[0012] The updating module is used to perform protection control on the subway DC power supply system according to the DC power supply system protection strategy, obtain the protection execution result, analyze the protection execution result, and update the DC power supply system protection strategy.

[0013] In the technical solution provided by the present invention, by real-time collection and preprocessing of the current data of the subway DC power supply system, accurate current information can be obtained in time, providing a reliable data basis for subsequent analysis. This real-time performance enables the system to respond quickly to current changes and improves the timeliness of protection. Secondly, the pre-processed current data is feature extracted and analyzed, and the obtained current characteristic parameters include current change rate, current peak value and harmonic content. These parameters fully reflect the dynamic characteristics and quality status of the current, making fault diagnosis more accurate. By performing fault diagnosis on the current characteristic parameters, the fault type judgment results obtained cover three states: normal operation, slight abnormality and serious fault. This hierarchical diagnosis method improves the system's ability to identify faults of different degrees, which is conducive to taking corresponding protection measures. A comprehensive analysis is performed on the fault type judgment results and the current operating state to obtain dynamic protection thresholds, including overcurrent protection thresholds and differential protection thresholds. This dynamic adjustment mechanism enables the protection threshold to adapt to the real-time state of the system, avoiding the problem of over-protection or under-protection caused by fixed thresholds. By analyzing the dynamic protection threshold and current characteristic parameters, the DC power supply system protection strategy obtained includes three levels: early warning, current limiting and disconnection protection. This multi-level protection strategy can take corresponding measures according to the severity of the fault, which not only ensures system safety but also minimizes unnecessary power outages. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 The figure is a flow chart of a circuit protection method based on real-time current data in an embodiment of the present invention.

[0016] Figure 2 Schematic diagram of a circuit protection system based on real-time current data in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. 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.

[0018] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0019] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0020] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 , Figure 1 is a flow chart of a circuit protection method based on real-time current data according to an embodiment of the present invention. Figure 1 As shown, the following steps are included:

[0021] S101, collecting and preprocessing the current data of the subway DC power supply system in real time to obtain preprocessed current data;

[0022] S102, extracting and analyzing the preprocessed current data to obtain current characteristic parameters;

[0023] S103, performing fault diagnosis on the current characteristic parameters to obtain a fault type judgment result, where the fault type judgment result includes normal operation, slight abnormality, and severe fault;

[0024] S104, comprehensively analyzing the fault type judgment result and the current operating status to obtain a dynamic protection threshold;

[0025] S105, analyzing the dynamic protection threshold and the current characteristic parameters to obtain a DC power supply system protection strategy, where the DC power supply system protection strategy includes early warning, current limiting and disconnection protection;

[0026] S106. Perform protection control on the subway DC power supply system according to the DC power supply system protection strategy, obtain protection execution results, analyze the protection execution results, and update the DC power supply system protection strategy.

[0027] Specifically, current data is collected in real time through a high-precision current sensor, and the collection frequency is set to once per millisecond to ensure that instantaneous current changes are captured. The collected raw data is digitally filtered and median filtered to remove high-frequency noise and outliers, and then normalized to obtain pre-processed current data. Feature extraction and analysis are performed on these data, and the first-order derivative (di / dt) of the current, that is, the current change rate, is calculated. The sliding window technology is used to calculate the current mean, maximum, minimum and standard deviation in a short period of time, and the harmonic components of the current are analyzed by fast Fourier transform (FFT) to obtain the current characteristic parameters. The machine learning algorithm is used to diagnose the faults of the current characteristic parameters, and the results are classified into normal operation, minor abnormalities and serious faults. The diagnosis process considers historical data and environmental factors such as temperature and humidity to improve accuracy. The protection threshold is dynamically calculated based on the diagnosis results and the current operating status. Fixed thresholds are no longer used, but adjusted according to real-time conditions. Considering factors such as the current load level, ambient temperature, fault diagnosis results and current change trends, a fuzzy logic controller is used for comprehensive analysis to output dynamic protection thresholds. Based on the dynamic protection threshold and current characteristic parameters, an adaptive protection strategy is formulated. Using a multi-level decision tree, the system decides to take measures such as early warning, current limiting or disconnection protection based on factors such as the degree, duration and rate of change of the current exceeding the threshold. The strategy formulation also takes into account the importance of the equipment and the power supply reliability requirements, and adopts a more conservative protection strategy for key equipment.

[0028] Finally, the protection strategy is executed and feedback is collected. The system sends control instructions to the relevant circuit breaker or current limiting device through the control interface to execute the protection action. After execution, the circuit status is continuously monitored and feedback information is collected, including whether the protection action is successfully executed, the circuit status after execution, whether there is a secondary fault, etc. The system evaluates the effect of the protection action. If it fails to solve the problem or causes new problems, the strategy is adjusted immediately and re-executed. All execution processes and feedback information are recorded in detail for subsequent system optimization and fault analysis.

[0029] Take a subway line as an example. During normal operation, the system collects current data every millisecond and obtains current characteristic parameters after preprocessing. Assume that at a certain moment, the system detects a sudden increase in current, and the di / dt value reaches 2000 A / s, which exceeds the normal operating range. The fault diagnosis algorithm determines this situation as a minor abnormality. The system then adjusts the protection threshold, reducing the overcurrent protection threshold from the original 3000 A to 2800 A, and adjusting the differential protection threshold from 500 A to 450 A. Based on these new thresholds, the system formulates a protection strategy that includes early warning and current limiting measures. After executing the strategy, the system monitors that the current growth rate begins to decline and eventually stabilizes within a safe range, avoiding potential serious faults. In this process, the system continuously updates the protection strategy to ensure the safe and stable operation of the power supply system.

[0030] By executing the above steps, the current data of the subway DC power supply system is collected and preprocessed in real time, and accurate current information can be obtained in time, providing a reliable data basis for subsequent analysis. This real-time performance enables the system to respond quickly to current changes and improves the timeliness of protection. Secondly, the preprocessed current data is feature extracted and analyzed, and the current characteristic parameters obtained include current change rate, current peak value and harmonic content. These parameters fully reflect the dynamic characteristics and quality status of the current, making fault diagnosis more accurate. By performing fault diagnosis on the current characteristic parameters, the fault type judgment results obtained cover three states: normal operation, slight abnormality and serious fault. This hierarchical diagnosis method improves the system's ability to identify faults of different degrees and is conducive to taking corresponding protection measures. A comprehensive analysis of the fault type judgment results and the current operating status is performed to obtain dynamic protection thresholds, including overcurrent protection thresholds and differential protection thresholds. This dynamic adjustment mechanism enables the protection threshold to adapt to the real-time state of the system, avoiding the problem of over-protection or under-protection caused by fixed thresholds. By analyzing the dynamic protection threshold and current characteristic parameters, the DC power supply system protection strategy obtained includes three levels: early warning, current limiting and disconnection protection. This multi-level protection strategy can take corresponding measures according to the severity of the fault, which not only ensures system safety but also minimizes unnecessary power outages.

[0031] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0032] (1) Perform high-frequency sampling on the output signal of the current sensor of the subway DC power supply system to obtain raw current data, and perform digital filtering on the raw current data to obtain filtered current data;

[0033] (2) normalizing the filtered current data to obtain normalized current data, and segmenting the normalized current data into time windows to obtain segmented current data;

[0034] (3) performing noise removal processing on the segmented current data to obtain noise-reduced current data, and performing outlier detection on the noise-reduced current data to obtain an outlier marker;

[0035] (4) performing interpolation processing on the noise-reduced current data according to the abnormal value marker to obtain interpolation-corrected current data, and performing smoothing processing on the interpolation-corrected current data to obtain smoothed current data;

[0036] (5) Extracting time domain features from the smoothed current data to obtain time domain feature parameters, and performing frequency domain transformation on the smoothed current data to obtain frequency domain data;

[0037] (6) Extract frequency domain features from the frequency domain data to obtain frequency domain feature parameters, and combine the time domain feature parameters with the frequency domain feature parameters to obtain preprocessed current data.

[0038] Specifically, in the subway DC power supply system, the implementation of the circuit protection method based on real-time current data first requires high-frequency sampling of the current sensor output signal. The sampling frequency is set to once per millisecond to ensure that the instantaneous current changes can be captured. The raw current data obtained by this high-frequency sampling contains rich information, but it may also contain various interferences and noises. Therefore, it is necessary to digitally filter the raw current data. Digital filtering uses a finite impulse response (FIR) filter, which can effectively remove high-frequency noise, retain useful signal components, and obtain filtered current data. Next, the filtered current data is normalized. Normalization maps the data to the [0, 1] interval, which helps to eliminate the influence of different dimensions and makes subsequent data processing more accurate. After normalization, the data is segmented into time windows, and the continuous data stream is segmented into several time periods, each of which contains a certain number of data points. This segmentation process is conducive to subsequent feature extraction and analysis.

[0039] The next key step is to remove noise from the segmented current data. The wavelet transform method is used here, which can analyze in both the time domain and the frequency domain and effectively remove the noise components in the data. The current data after noise reduction is smoother, but there may still be outliers. Therefore, the triple standard deviation method is used to detect outliers on the noise-reduced data to obtain outlier markers. According to the outlier markers, the noise-reduced current data is interpolated. For data points marked as abnormal, linear interpolation is used for correction. This method can effectively repair the outliers in the data and maintain the continuity of the data. The interpolated and corrected data is then smoothed, and the moving average method is used to further eliminate small fluctuations in the data to obtain smoothed current data.

[0040] Extracting time domain features from smoothed current data is an important data processing step. Time domain features include statistical features such as mean, variance, peak, and valley, which can reflect the basic statistical characteristics of current data. At the same time, the smoothed current data is transformed into frequency domain, and the fast Fourier transform (FFT) method is used to convert time domain data into frequency domain data. Frequency domain data can reflect the frequency composition of the current signal and help identify periodic changes and harmonic components. Finally, frequency domain features are extracted from the frequency domain data, including information such as the amplitude and phase of the main frequency components. The time domain feature parameters and frequency domain feature parameters are combined to obtain comprehensive preprocessed current data. These data contain the time domain and frequency domain characteristics of the current signal, providing comprehensive data support for subsequent fault diagnosis and protection strategy formulation.

[0041] For example, in the DC power supply system of a subway line, the current sensor collects data once every millisecond, and a total of 10,000 data points are collected within 10 seconds. There is 50Hz power frequency interference and random noise in the original data. After removing the high-frequency noise through the FIR filter, the data fluctuation is significantly reduced. Normalization processing maps the current value to between 0 and 1, which is convenient for subsequent processing. The time window segmentation uses 1 second (1000 data points) as a window. After wavelet transform noise reduction, the signal-to-noise ratio is improved by about 5dB. Outlier detection found 5 abnormal points, which were corrected by linear interpolation. Moving average smoothing uses a window of 50 points. Time domain feature extraction obtains statistics such as a mean of 0.6 and a standard deviation of 0.1. After FFT transformation, it is found that there is an obvious 300Hz harmonic component with an amplitude of 5% of the fundamental wave.

[0042] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0043] (1) Perform time series analysis on the preprocessed current data to obtain the current time series characteristics, and perform differential operation on the current time series characteristics to obtain the current change rate;

[0044] (2) Perform extreme value detection on the preprocessed current data to obtain a set of current extreme value points, and extract the maximum value from the set of current extreme value points to obtain the current peak value;

[0045] (3) Performing fast Fourier transform on the preprocessed current data to obtain current spectrum data, and performing harmonic analysis on the current spectrum data to obtain harmonic content;

[0046] (4) Statistical feature extraction is performed on the preprocessed current data to obtain current statistical feature parameters, and the current statistical feature parameters are combined with the current change rate, current peak value and harmonic content to obtain a current feature vector;

[0047] (5) Performing dimensionality reduction processing on the current characteristic vector to obtain current characteristic parameters after dimensionality reduction, and performing standardization processing on the current characteristic parameters after dimensionality reduction to obtain current characteristic parameters.

[0048] Specifically, the preprocessed current data is subjected to time series analysis, and the autoregressive moving average (ARMA) model is used to extract the current time series features. The ARMA model can be expressed as:

[0049]

[0050] in, yes The current value at the moment, is time The current value before time units, is a constant term, is the autoregressive coefficient, is the moving average coefficient, yes White noise at all times, is time time units ago, is the order of the moving average term. By fitting the ARMA model, the time series characteristics of the current data can be obtained. Subsequently, the current time series characteristics are differentiated to obtain the current change rate:

[0051]

[0052] in, express The current value at the moment, represents the rate of change of current, is the time interval.

[0053] Next, the preprocessed current data is subjected to extreme value detection, and the sliding window method is used to identify the local maximum and minimum values. The window size is set to , for each point in the time series , if it satisfies:

[0054]

[0055] but It is identified as a local maximum. Similarly, the local minimum can be identified. The global maximum is extracted from the obtained current extreme point set, which is the current peak value. In order to analyze the frequency characteristics of the current data, the pre-processed current data is subjected to a fast Fourier transform (FFT). FFT converts the time domain signal into a frequency domain representation, performs harmonic analysis on the obtained current spectrum data, and calculates the total harmonic distortion (THD):

[0056]

[0057] in, is the amplitude of the nth harmonic, is the fundamental amplitude. At the same time, the statistical features of the preprocessed current data are extracted, including mean, standard deviation, skewness and kurtosis. These statistical features, together with the current change rate, current peak value and harmonic content obtained previously, constitute the current feature vector.

[0058] In order to reduce the data dimension and improve the processing efficiency, the current feature vector is subjected to principal component analysis (PCA) dimensionality reduction. PCA transforms the original features into linearly independent new features through orthogonal transformation, and selects the first k principal components as the features after dimensionality reduction. Finally, the current characteristic parameters after dimensionality reduction are standardized to obtain the current characteristic parameters.

[0059] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0060] (1) Performing data normalization processing on the current characteristic parameters to obtain normalized current characteristic parameters, and performing cluster analysis on the normalized current characteristic parameters to obtain current characteristic clustering results;

[0061] (2) Based on the current feature clustering results, the normalized current feature parameters are divided into thresholds to obtain the preliminary fault type, and the preliminary fault type is fuzzy evaluated to obtain the fuzzy fault level;

[0062] (3) Defuzzifying the fuzzy fault level to obtain a fault severity index, and correcting the preliminary fault type according to the fault severity index to obtain a corrected fault type;

[0063] (4) Performing a time series analysis on the corrected fault type to obtain the fault development trend, and dynamically adjusting the corrected fault type according to the fault development trend to obtain a dynamic fault type;

[0064] (5) Comprehensively evaluate the dynamic fault types and obtain the fault type judgment results, which include normal operation, minor abnormalities and serious faults.

[0065] Specifically, the current characteristic parameters are normalized, and the minimum-maximum normalization method is used to map each characteristic parameter to the interval [0,1] to obtain the normalized current characteristic parameters. This ensures that characteristic parameters of different dimensions can be compared and analyzed. Next, the normalized current characteristic parameters are clustered and analyzed, and the K-means clustering algorithm is used to divide the data points into multiple clusters. The K-means algorithm minimizes the sum of squares of the distance from each data point to the center of its cluster through iterative optimization, thereby obtaining the current characteristic clustering results. According to the current characteristic clustering results, the normalized current characteristic parameters are thresholded. The process of setting the threshold takes into account the location of the cluster center and the variance within the cluster, and divides the data points into different areas corresponding to different preliminary fault types. Subsequently, the preliminary fault type is fuzzily evaluated and a fuzzy logic system is introduced. The fuzzy logic system includes three steps: fuzzification, reasoning, and defuzzification. In the fuzzification process, the preliminary fault type is converted into the membership degree of the fuzzy set to obtain the fuzzy fault level.

[0066] The fuzzy fault level is defuzzified, and the center of gravity method is used to convert the fuzzy quantity into a clear numerical value to obtain the fault severity index. According to the fault severity index, the preliminary fault type is corrected to obtain the corrected fault type. The result of fuzzy evaluation is considered to make the fault type judgment more accurate and reliable. Next, the corrected fault type is analyzed in time series, and the sliding time window method is used to analyze the change trend of the fault type within a certain time range to obtain the fault development trend. According to the fault development trend, the corrected fault type is dynamically adjusted to obtain the dynamic fault type. The time evolution characteristics of the fault are considered to reflect the changes in the fault state in a timely manner. Finally, the dynamic fault type is comprehensively evaluated, and the fault type, severity and development trend are comprehensively considered to obtain the final fault type judgment result. The fault type judgment results include three categories: normal operation, minor abnormality and serious fault, which provide a basis for the subsequent protection strategy formulation.

[0067] For example, in the DC power supply system of a subway line, current data within 10 minutes were collected, sampled once per second, and a total of 600 data points. The extracted current characteristic parameters include current mean, standard deviation, peak factor, and harmonic content. After the minimum-maximum normalization of these characteristic parameters, the normalized current characteristic parameters were obtained. The K-means algorithm was used for cluster analysis, and the number of clusters K=3 was set. After 50 iterations, a stable clustering result was obtained. According to the clustering results, the threshold was set to divide the data points into three preliminary fault types. In the fuzzy evaluation stage, three fuzzy sets of "low", "medium", and "high" were defined, and the preliminary fault types were converted into fuzzy fault levels through membership functions. For example, the membership of a data point in the three fuzzy sets of "low", "medium", and "high" were 0.2, 0.7, and 0.1, respectively. The centroid method was used for defuzzification, and the fault severity index was 0.45 (assuming that the range of 0-1 corresponds to normal to severe faults).

[0068] The timing analysis uses a 5-minute sliding time window to analyze the changing trend of the fault type. For example, within a certain time window, the proportion of minor anomalies increases from 10% to 30%, indicating that the fault has a tendency to worsen. According to this trend, the fault type is dynamically adjusted. Finally, the comprehensive evaluation takes into account the fault type, severity index and development trend. In this case, although the current fault severity index is 0.45, which belongs to the range of minor anomalies, considering the trend of worsening faults, the final judgment result may be adjusted from minor anomalies to serious faults, so that corresponding protection measures can be taken in time. This method of dynamic adjustment and comprehensive evaluation can more accurately reflect the actual operating status of the subway DC power supply system and provide a reliable judgment basis for circuit protection.

[0069] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0070] (1) Weights are assigned to the fault type judgment results to obtain the fault type weight coefficient, and the current operating state is evaluated according to the fault type weight coefficient to obtain the operating state evaluation value;

[0071] (2) Perform fuzzy reasoning on the operating status evaluation value to obtain a fuzzy protection threshold, and defuzzify the fuzzy protection threshold to obtain an initial dynamic protection threshold;

[0072] (3) Adaptively adjusting the initial dynamic protection threshold according to the historical protection threshold data to obtain the adjusted dynamic protection threshold, and classifying the adjusted dynamic protection threshold to obtain the overcurrent protection threshold and the differential protection threshold;

[0073] (4) Analyze the time characteristics of the overcurrent protection threshold to obtain the overcurrent protection time curve, and perform sensitivity analysis on the differential protection threshold to obtain the differential protection sensitivity coefficient;

[0074] (5) According to the overcurrent protection time curve and the differential protection sensitivity coefficient, the overcurrent protection threshold and the differential protection threshold are optimized to obtain the dynamic protection threshold.

[0075] Specifically, in the circuit protection method of the subway DC power supply system, weighting the fault type judgment results is a key step in determining the dynamic protection threshold. First, according to the severity and frequency of the fault type, the fault type judgment results are weighted to obtain the fault type weight coefficient. The weight allocation adopts the hierarchical analysis method (AHP) to obtain the weight of each fault type by constructing a judgment matrix and calculating the eigenvector. The calculation formula of the weight coefficient is:

[0076]

[0077] in, It is The weight of each fault type, is the first Line The elements of the column, n is the number of fault types, is the first Line Elements of a column.

[0078] According to the fault type weight coefficient, the current operating status is evaluated to obtain the operating status evaluation value. The evaluation value is calculated using the weighted sum method:

[0079]

[0080] in, is the operating status evaluation value, is the severity score of the ith fault type.

[0081] Perform fuzzy reasoning on the operating status evaluation value to obtain the fuzzy protection threshold. The fuzzy reasoning adopts the Mamdani reasoning method. By setting the fuzzy rule base and membership function, the operating status evaluation value is mapped to the fuzzy protection threshold. Then, the fuzzy protection threshold is defuzzified and the initial dynamic protection threshold is obtained by the centroid method. The initial dynamic protection threshold is adaptively adjusted according to the historical protection threshold data to obtain the adjusted dynamic protection threshold. The adaptive adjustment adopts the exponentially weighted moving average (EWMA) method:

[0082]

[0083] in, is the adjusted threshold at time t, is the threshold at time t-1, is the initial dynamic protection threshold, is the smoothing coefficient (0< <1). The adjusted dynamic protection threshold is classified to obtain the overcurrent protection threshold and the differential protection threshold. The classification process takes into account the characteristics and requirements of different protection types.

[0084] The time characteristic analysis of the overcurrent protection threshold is performed to obtain the overcurrent protection time curve. The time characteristic analysis uses the inverse time characteristic curve:

[0085]

[0086] Where m is the operating time, I is the fault current, is the starting current, , and L are adjustable parameters. A sensitivity analysis is performed on the differential protection threshold to obtain the differential protection sensitivity coefficient. The sensitivity analysis takes into account the current transformer error and the differential current during normal system operation, and the sensitivity coefficient is determined by calculating the minimum detectable fault current.

[0087] Finally, the overcurrent protection threshold and differential protection threshold are optimized according to the overcurrent protection time curve and the differential protection sensitivity coefficient to obtain the final dynamic protection threshold. The optimization process adopts a multi-objective optimization algorithm, taking into account the sensitivity, selectivity and reliability of the protection.

[0088] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0089] (1) Compare and analyze the dynamic protection threshold and current characteristic parameters to obtain the limit-crossing degree index, and conduct risk level assessment based on the limit-crossing degree index to obtain the protection risk level;

[0090] (2) The protection risk levels are graded to obtain a graded protection strategy, and the protection priorities of various components of the DC power supply system are sorted according to the graded protection strategy to obtain a protection execution order;

[0091] (3) Arrange the hierarchical protection strategy in time sequence according to the protection execution order to obtain a time sequence protection plan, and conduct a feasibility analysis on the time sequence protection plan to obtain a feasibility evaluation result;

[0092] (4) Adjust the sequential protection scheme according to the feasibility assessment results to obtain an adjusted protection scheme, and refine the adjusted protection scheme to obtain specific protection operation instructions;

[0093] (5) The specific protection operation instructions are divided into three categories: early warning, current limiting and disconnection protection, and the protection strategy of the DC power supply system is obtained.

[0094] Specifically, the difference between the current characteristic parameter and the dynamic protection threshold is calculated and normalized to obtain the over-limit index. This index reflects the degree to which the current current state exceeds the protection threshold. According to the over-limit index, the fuzzy evaluation method is used to evaluate the risk level and obtain the protection risk level. The fuzzy evaluation method maps the over-limit index to different risk levels, such as low risk, medium risk and high risk, by establishing a fuzzy relationship matrix. The protection risk level is graded and the hierarchical protection strategy is obtained using the decision tree algorithm. The construction of the decision tree is based on historical data and expert knowledge, and the protection measures to be taken under different risk levels are considered. According to the hierarchical protection strategy, the protection priority of each component of the DC power supply system is sorted. The sorting process uses the analytic hierarchy process (AHP), considering factors such as the importance of the component, the scope of the fault impact and the difficulty of repair, and finally obtains the protection execution order.

[0095] According to the protection execution order, the hierarchical protection strategy is scheduled in time to obtain a sequential protection plan. The critical path method (CPM) is used for scheduling to determine the execution time and order of each protection measure to maximize the protection effect and minimize the system interruption time. The feasibility analysis of the sequential protection plan is carried out. The Monte Carlo simulation method is used to evaluate the reliability and effectiveness of the plan through multiple random simulations of different fault scenarios to obtain the feasibility evaluation results. According to the feasibility evaluation results, the sequential protection plan is adjusted. The adjustment process uses a genetic algorithm to find a balance between the protection effect and the execution cost through iterative optimization to obtain the adjusted protection plan. The adjusted protection plan is refined, and each protection measure is decomposed into specific operation steps, including equipment operation instructions, parameter settings, and manual intervention, to obtain specific protection operation instructions.

[0096] Finally, the specific protection operation instructions are divided into three categories: early warning, current limiting and disconnection protection. Early warning instructions include sending alarm information, starting backup equipment, etc.; current limiting instructions include adjusting rectifier output, activating dynamic current limiting devices, etc.; disconnection protection instructions include tripping circuit breakers, isolating fault sections, etc. This classification method forms a hierarchical DC power supply system protection strategy, which can take corresponding protection measures according to the severity of the fault.

[0097] For example, in the DC power supply system of a subway line, a sudden increase in feeder current was monitored in real time. The current characteristic parameters show that the current value is 3000A, the current change rate is 1000A / s, and the harmonic content is 5%. Compared with the dynamic protection threshold (current threshold 2500A, change rate threshold 800A / s, harmonic content threshold 3%), the calculated over-limit degree indicators are 0.2, 0.25 and 0.67 respectively. Through the fuzzy evaluation method, the comprehensive evaluation results show that the protection risk level is "medium risk". Based on this risk level, the hierarchical protection strategy generated by the decision tree algorithm includes: starting the backup power supply, limiting the traction power and preparing for the circuit breaker to trip. The AHP method is used to prioritize the system components, and the protection execution order is: rectifier, circuit breaker, feeder. The CPM method is used for timing arrangement, and the timing protection scheme is: starting the backup power supply at T0, limiting the traction power at T0+5s, and preparing for the circuit breaker to trip at T0+10s. Monte Carlo simulations were performed 1,000 times for different failure scenarios, and the feasibility of the solution was evaluated to be 92%.

[0098] Through genetic algorithm optimization, the adjusted protection scheme will advance the time of limiting traction power to T0+3s, and add an intermediate step: local current limiting at T0+7s. The detailed protection operation instructions include: sending early warning information to the control center, limiting the rectifier output current to 2800A, adjusting the tap of the traction transformer to 85%, preparing for the tripping of circuit breaker No. 3, etc. These instructions are divided into three categories: early warning (sending early warning information), current limiting (limiting rectifier output, adjusting transformer taps), and disconnection protection (preparing for circuit breaker tripping).

[0099] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0100] (1) Analyze the protection strategy of the DC power supply system to obtain specific execution instructions, and generate a control signal sequence according to the specific execution instructions to obtain a protection control signal;

[0101] (2) Sending protection control signals to relevant equipment of the subway DC power supply system, executing protection operations, obtaining real-time execution status, and monitoring the real-time execution status to obtain protection execution results;

[0102] (3) Conduct statistical analysis on the protection execution results to obtain protection effect evaluation indicators, and evaluate the effectiveness of the DC power supply system protection strategy based on the protection effect evaluation indicators to obtain strategy evaluation results;

[0103] (4) Adjusting the protection strategy of the DC power supply system according to the strategy evaluation results to obtain an adjusted protection strategy;

[0104] (5) Compare and analyze the adjusted protection strategy with the historical protection strategy to obtain the strategy improvement effect, and determine the updated DC power supply system protection strategy based on the strategy improvement effect.

[0105] Specifically, the semantic analysis technology is used to parse the protection strategy and convert the high-level strategy description into specific execution instructions. This process includes instruction decomposition, parameter extraction and sequence determination. According to the specific execution instructions obtained by the analysis, the control signal sequence is generated through the instruction mapping algorithm. This algorithm maps each execution instruction to the corresponding device control command, and considers the execution timing and interdependence, and finally obtains a series of orderly protection control signals. Next, the protection control signal is sent to the relevant equipment of the subway DC power supply system through the communication network. The IEC 61850 communication protocol is used here to ensure the reliable transmission and rapid response of the control signal. After receiving the control signal, the device performs the corresponding protection operation, such as adjusting the rectifier output, triggering the circuit breaker action, etc. At the same time, the execution status is monitored by the real-time data acquisition system, including the equipment action time, current change trend, etc., to obtain the real-time execution status. The real-time execution status is data fused and analyzed, and the Kalman filter algorithm is used to eliminate the noise influence to obtain accurate protection execution results.

[0106] The protection execution results are statistically analyzed and a multi-dimensional evaluation method is adopted. This includes calculating the protection action time, current suppression effect, system stability index, etc. The principal component analysis (PCA) method is used to integrate multiple evaluation indicators into several key protection effect evaluation indicators. Based on these evaluation indicators, the fuzzy comprehensive evaluation method is used to evaluate the effectiveness of the DC power supply system protection strategy to obtain the strategy evaluation results. The evaluation process takes into account multiple aspects such as protection speed, reliability, and selectivity. According to the strategy evaluation results, the DC power supply system protection strategy is adjusted using an adaptive adjustment algorithm. This algorithm is based on the principle of reinforcement learning and finds better protection strategy parameters through continuous trial and error and optimization. The adjustment process takes into account the historical execution effect, the current system status, and possible future fault scenarios to obtain the adjusted protection strategy.

[0107] Finally, the adjusted protection strategy is compared with the historical protection strategy. Time series analysis methods, such as the autoregressive integrated moving average (ARIMA) model, are used to model and predict the changing trend of strategy parameters. The effect of strategy improvement is obtained by calculating the effect indicators of strategy improvement, such as the percentage of protection reliability improvement and the average response time reduction. According to the effect of strategy improvement, the decision tree algorithm is used to determine whether to adopt the new strategy, so as to obtain the updated DC power supply system protection strategy.

[0108] For example, a DC power supply system of a subway line detected a sudden increase in feeder current during operation. The specific execution instructions obtained after the protection strategy analysis include: limiting the rectifier output current to 2800A, triggering the fast circuit breaker, and starting the backup power supply. These instructions are converted into control signal sequences such as "SET_RECTIFIER_CURRENT2800", "TRIP_BREAKER 3", and "START_BACKUP_POWER" through the instruction mapping algorithm. These control signals are sent to related devices through the IEC 61850 protocol. During the execution process, real-time monitoring shows that the rectifier current drops to 2800A within 50ms, the circuit breaker completes tripping within 30ms, and the backup power supply starts within 2s. After Kalman filtering, accurate execution time and current change curves are obtained.

[0109] Data statistical analysis shows that the average response time of this protection action is 40ms, the current suppression effect reaches 95%, and the system stability index remains at 0.98. Through PCA analysis, these indicators are combined into a protection effect evaluation index of 0.92 (full score is 1). The fuzzy comprehensive evaluation results show that the effectiveness score of the current strategy is 0.88. Based on this evaluation result, the adaptive adjustment algorithm recommends adjusting the rectifier current limiting threshold to 2750A and the circuit breaker triggering time to 5ms earlier. The adjusted strategy is compared with historical data through the ARIMA model, predicting that in similar fault conditions in the future, the average response time can be reduced to 35ms and the current suppression effect can be improved to 97%. The decision tree algorithm comprehensively considers these improvements and decides to adopt a new protection strategy. The updated DC power supply system protection strategy not only includes new parameter settings, but also adds a new protection measure: when a current surge is detected, the traction power supply voltage is automatically reduced by 5% to further improve the protection effect.

[0110] The embodiment of the present invention also provides a circuit protection system based on real-time current data, such as Figure 2 As shown, the circuit protection system based on real-time current data specifically includes:

[0111] The processing module 201 is used to collect and preprocess the current data of the subway DC power supply system in real time to obtain the preprocessed current data;

[0112] An extraction module 202 is used to extract and analyze the preprocessed current data to obtain current characteristic parameters;

[0113] A diagnosis module 203, used to perform fault diagnosis on the current characteristic parameters to obtain a fault type judgment result, wherein the fault type judgment result includes normal operation, slight abnormality and serious fault;

[0114] The analysis module 204 is used to comprehensively analyze the fault type judgment result and the current operating state to obtain a dynamic protection threshold;

[0115] A protection module 205, configured to analyze the dynamic protection threshold and the current characteristic parameter to obtain a DC power supply system protection strategy, wherein the DC power supply system protection strategy includes early warning, current limiting and disconnection protection;

[0116] The updating module 206 is used to perform protection control on the subway DC power supply system according to the DC power supply system protection strategy, obtain a protection execution result, analyze the protection execution result, and update the DC power supply system protection strategy.

[0117] Through the collaborative work of the above modules, by real-time collection and preprocessing of the current data of the subway DC power supply system, accurate current information can be obtained in time, providing a reliable data basis for subsequent analysis. This real-time performance enables the system to respond quickly to current changes and improves the timeliness of protection. Secondly, the preprocessed current data is feature extracted and analyzed, and the current characteristic parameters obtained include current change rate, current peak value and harmonic content. These parameters fully reflect the dynamic characteristics and quality status of the current, making fault diagnosis more accurate. By performing fault diagnosis on the current characteristic parameters, the fault type judgment results obtained cover three states: normal operation, slight abnormality and serious fault. This hierarchical diagnosis method improves the system's ability to identify faults of different degrees and is conducive to taking corresponding protection measures. A comprehensive analysis of the fault type judgment results and the current operating status is performed to obtain dynamic protection thresholds, including overcurrent protection thresholds and differential protection thresholds. This dynamic adjustment mechanism enables the protection threshold to adapt to the real-time state of the system, avoiding the problem of over-protection or under-protection caused by fixed thresholds. By analyzing the dynamic protection threshold and current characteristic parameters, the DC power supply system protection strategy obtained includes three levels: early warning, current limiting and disconnection protection. This multi-level protection strategy can take corresponding measures according to the severity of the fault, which not only ensures system safety but also minimizes unnecessary power outages.

[0118] The above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. Although the present invention has been described in detail with reference to the embodiments, a person skilled in the art should understand that the specific implementation modes of the present invention can still be modified or replaced by equivalents, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A circuit protection method based on real-time current data, characterized in that: include: The current data of the subway DC power supply system is collected and preprocessed in real time to obtain the preprocessed current data; Extracting and analyzing the preprocessed current data to obtain current characteristic parameters; Performing fault diagnosis on the current characteristic parameters to obtain a fault type judgment result, wherein the fault type judgment result includes normal operation, slight abnormality and serious fault; Comprehensively analyzing the fault type judgment result and the current operating status to obtain a dynamic protection threshold; Analyzing the dynamic protection threshold and the current characteristic parameter to obtain a DC power supply system protection strategy, wherein the DC power supply system protection strategy includes early warning, current limiting and disconnection protection; The subway DC power supply system is protected and controlled according to the DC power supply system protection strategy to obtain a protection execution result, analyze the protection execution result, and update the DC power supply system protection strategy.

2. The circuit protection method based on real-time current data according to claim 1, characterized in that: The step of collecting and preprocessing the current data of the subway DC power supply system in real time to obtain the preprocessed current data includes: Performing high-frequency sampling on the output signal of the current sensor of the subway DC power supply system to obtain raw current data, and performing digital filtering on the raw current data to obtain filtered current data; Normalizing the filtered current data to obtain normalized current data, and segmenting the normalized current data into time windows to obtain segmented current data; Performing noise removal processing on the segmented current data to obtain noise-reduced current data, and performing outlier detection on the noise-reduced current data to obtain an outlier mark; performing interpolation processing on the noise-reduced current data according to the abnormal value mark to obtain interpolation-corrected current data, and performing smoothing processing on the interpolation-corrected current data to obtain smoothed current data; Performing time domain feature extraction on the smoothed current data to obtain time domain feature parameters, and performing frequency domain transformation on the smoothed current data to obtain frequency domain data; Frequency domain features are extracted from the frequency domain data to obtain frequency domain feature parameters, and the time domain feature parameters and the frequency domain feature parameters are combined to obtain the preprocessed current data.

3. The circuit protection method based on real-time current data according to claim 1, characterized in that: The step of extracting and analyzing the preprocessed current data to obtain current characteristic parameters includes: Performing time series analysis on the preprocessed current data to obtain current time series characteristics, and performing differential operation on the current time series characteristics to obtain current change rate; Performing extreme value detection on the preprocessed current data to obtain a current extreme value point set, and extracting a maximum value from the current extreme value point set to obtain a current peak value; Performing a fast Fourier transform on the preprocessed current data to obtain current spectrum data, and performing a harmonic analysis on the current spectrum data to obtain harmonic content; Performing statistical feature extraction on the preprocessed current data to obtain current statistical feature parameters, and combining the current statistical feature parameters with the current change rate, the current peak value and the harmonic content to obtain a current feature vector; The current characteristic vector is subjected to dimensionality reduction processing to obtain the current characteristic parameters after dimensionality reduction, and the current characteristic parameters after dimensionality reduction are subjected to standardization processing to obtain the current characteristic parameters.

4. The circuit protection method based on real-time current data according to claim 1, characterized in that: The step of performing fault diagnosis on the current characteristic parameters to obtain a fault type judgment result, wherein the fault type judgment result includes normal operation, slight abnormality and serious fault, comprises: Performing data normalization processing on the current characteristic parameters to obtain normalized current characteristic parameters, and performing cluster analysis on the normalized current characteristic parameters to obtain current characteristic clustering results; Performing threshold division on the normalized current characteristic parameters according to the current characteristic clustering result to obtain a preliminary fault type, and performing fuzzy evaluation on the preliminary fault type to obtain a fuzzy fault level; Defuzzifying the fuzzy fault level to obtain a fault severity index, and correcting the preliminary fault type according to the fault severity index to obtain a corrected fault type; Performing a time series analysis on the corrected fault type to obtain a fault development trend, and dynamically adjusting the corrected fault type according to the fault development trend to obtain a dynamic fault type; A comprehensive evaluation is performed on the dynamic fault type to obtain the fault type judgment result, which includes normal operation, slight abnormality and serious fault.

5. The circuit protection method based on real-time current data according to claim 1, characterized in that: The step of comprehensively analyzing the fault type judgment result and the current operating state to obtain a dynamic protection threshold value includes: Performing weight allocation on the fault type judgment result to obtain a fault type weight coefficient, and evaluating the current operating state according to the fault type weight coefficient to obtain an operating state evaluation value; Performing fuzzy reasoning on the operation status evaluation value to obtain a fuzzy protection threshold, and performing defuzzification processing on the fuzzy protection threshold to obtain an initial dynamic protection threshold; Adaptively adjusting the initial dynamic protection threshold according to the historical protection threshold data to obtain an adjusted dynamic protection threshold, and classifying the adjusted dynamic protection threshold to obtain an overcurrent protection threshold and a differential protection threshold; Performing a time characteristic analysis on the overcurrent protection threshold to obtain an overcurrent protection time curve, and performing a sensitivity analysis on the differential protection threshold to obtain a differential protection sensitivity coefficient; The overcurrent protection threshold and the differential protection threshold are optimized according to the overcurrent protection time curve and the differential protection sensitivity coefficient to obtain the dynamic protection threshold.

6. The circuit protection method based on real-time current data according to claim 1, characterized in that: The dynamic protection threshold and the current characteristic parameter are analyzed to obtain a DC power supply system protection strategy, wherein the DC power supply system protection strategy includes early warning, current limiting and disconnection protection steps, including: Comparing and analyzing the dynamic protection threshold and the current characteristic parameter to obtain an over-limit degree index, and performing a risk level assessment based on the over-limit degree index to obtain a protection risk level; The protection risk levels are graded to obtain a graded protection strategy, and the protection priorities of the components of the DC power supply system are sorted according to the graded protection strategy to obtain a protection execution order; Performing a time sequence arrangement on the hierarchical protection strategy according to the protection execution order to obtain a time sequence protection scheme, and performing a feasibility analysis on the time sequence protection scheme to obtain a feasibility evaluation result; Adjusting the sequential protection scheme according to the feasibility assessment result to obtain an adjusted protection scheme, and refining the adjusted protection scheme to obtain specific protection operation instructions; The specific protection operation instructions are divided into three categories: early warning, current limiting and disconnection protection, to obtain the DC power supply system protection strategy.

7. The circuit protection method based on real-time current data according to claim 6, characterized in that: The step of performing protection control on the subway DC power supply system according to the DC power supply system protection strategy, obtaining a protection execution result, analyzing the protection execution result, and updating the DC power supply system protection strategy includes: Parsing the DC power supply system protection strategy to obtain specific execution instructions, and generating a control signal sequence according to the specific execution instructions to obtain a protection control signal; The protection control signal is sent to relevant equipment of the subway DC power supply system, a protection operation is performed, a real-time execution status is obtained, and the real-time execution status is monitored to obtain a protection execution result; Performing data statistical analysis on the protection execution results to obtain protection effect evaluation indicators, and evaluating the effectiveness of the DC power supply system protection strategy based on the protection effect evaluation indicators to obtain a strategy evaluation result; Adjusting the DC power supply system protection strategy according to the strategy evaluation result to obtain an adjusted protection strategy; The adjusted protection strategy is compared and analyzed with the historical protection strategy to obtain the strategy improvement effect, and an updated DC power supply system protection strategy is determined based on the strategy improvement effect.

8. A circuit protection system based on real-time current data, used to execute the circuit protection method based on real-time current data according to any one of claims 1 to 7, characterized in that: include: A processing module is used to collect and preprocess the current data of the subway DC power supply system in real time to obtain the preprocessed current data; An extraction module, used for performing feature extraction and analysis on the preprocessed current data to obtain current characteristic parameters; A diagnosis module, used to perform fault diagnosis on the current characteristic parameters to obtain a fault type judgment result, wherein the fault type judgment result includes normal operation, slight abnormality and serious fault; An analysis module is used to comprehensively analyze the fault type judgment result and the current operating state to obtain a dynamic protection threshold; A protection module, used for analyzing the dynamic protection threshold and the current characteristic parameter to obtain a DC power supply system protection strategy, wherein the DC power supply system protection strategy includes early warning, current limiting and disconnection protection; The updating module is used to perform protection control on the subway DC power supply system according to the DC power supply system protection strategy, obtain the protection execution result, analyze the protection execution result, and update the DC power supply system protection strategy.

Citation Information

Patent Citations

  • Real-time predictive systems for intelligent energy monitoring and management of electrical power networks

    CA2882796A1

  • Multi-source disturbance characteristic discrimination method and system based on time-frequency characteristic of power transmission line

    CN118707245A

  • Voltage fluctuation control method and system of subway direct current power supply system

    CN118713034A

  • Situation awareness and dynamic ensemble forecasting of abnormal behavior in cyber-physical system

    US20200067969A1