A method for identifying line contact faults based on AC / DC systems

Abstract

The present invention discloses a line contact fault identification method based on an AC / DC system, which relates to the technical field of power systems and comprises the following steps: collecting current data and voltage data of a target line; if the current current or voltage data exceeds a historical benchmark value, dynamically adjusting the sampling frequency and recording the current detection data and voltage detection data of the target line; analyzing the current transient abnormality index and the voltage transient abnormality index of the target line based on the current and voltage detection data; calculating the probability of occurrence of a line contact fault of the target line by comprehensively analyzing the current and voltage transient abnormality indexes and environmental assessment factors; determining whether the line contact fault of the target line is a permanent or transient line contact fault by statistically testing the current and voltage transient abnormality indexes of multiple subsequent sampling time intervals of the target line; realizing fault identification through multi-dimensional comprehensive analysis, and improving the accuracy of line contact fault identification and the scientific nature of line contact fault classification.

Description

Technical Field

[0001] The present invention belongs to the technical field of power systems and relates to a line contact fault identification method based on AC and DC systems. Background Art

[0002] In AC / DC system operation, line contact faults pose a critical threat to power system safety. Due to the complex structure, variable operating conditions, and susceptibility to environmental factors of AC / DC systems, traditional fault identification methods struggle to accurately and promptly detect line contact faults. This can not only damage equipment but also impact the stability of power supply. Therefore, a line contact fault identification method based on AC / DC systems is urgently needed to accurately detect and distinguish line contact fault types, ensuring the safe operation of power systems.

[0003] In the prior art, there are also some related solutions related to line contact fault identification. For example, the AC / DC line contact fault line selection method, device and computer-readable storage medium with Chinese patent publication number CN110794259A analyzes the first and second power directions of the busbar and determines the DC line where the AC / DC line contact fault occurs based on this, thereby achieving accurate fault line selection and reducing the scope of the fault impact.

[0004] Another Chinese patent, CN117233471A, is a method for detecting tree branch-to-line faults in medium-voltage distribution networks based on the gradual change characteristics of contact impedance. It fits the fault impedance by establishing a double exponential model, and determines whether a line-to-line fault has occurred by analyzing the fitting effect and fitting parameters. This reduces operation and maintenance costs, and solves the problem that existing fault detection technologies consume a lot of manpower and material resources and are unable to detect faults in a timely and effective manner.

[0005] Although the above schemes propose some solutions to line-touching fault identification, the existing technology still has the following limitations, specifically: (1) The existing technology usually uses a fixed sampling frequency to monitor the line current and voltage, and cannot dynamically adjust the sampling frequency according to the real-time operating status of the target line. The fixed sampling frequency will make it difficult to quickly capture the key characteristic signals in the early stage of the line-touching fault, thereby affecting the timeliness and accuracy of fault detection.

[0006] (2) In the process of identifying line-touching faults, the existing technology only identifies line-touching faults through the statistics of basic electrical parameters such as current and voltage, but does not analyze deep features such as the degree of current mutation and harmonic distortion, the duty cycle of high-level voltage and the degree of amplitude change. In addition, the existing technology does not take into account the influence of environmental parameters in the process of identifying line-touching faults, resulting in limited accuracy of fault identification and easy misjudgment or missed judgment.

[0007] (3) The existing technology does not classify line contact faults based on their duration, but rather classifies them based on line topology type. It lacks continuous tracking and statistical testing of transient abnormal indicators within the sampling time interval, and cannot determine the urgency of the line contact fault, which in turn affects the timeliness and accuracy of the fault handling process. Summary of the Invention

[0008] In view of this, in order to solve the problems raised in the above background technology, a line contact fault identification method based on AC / DC system is proposed.

[0009] The purpose of the present invention can be achieved through the following technical solutions: The present invention provides a line contact fault identification method based on AC / DC system, including: S1. Select a single line between adjacent detection points as the target line, and collect the current data and voltage data of the target line at a preset sampling frequency.

[0010] S2. Compare with historical benchmark values. If the current or voltage of the target line exceeds the limit, dynamically adjust the sampling frequency within the detection sampling time interval, and synchronously collect the current data and voltage data after the sampling frequency adjustment as the current detection data and voltage detection data of the target line.

[0011] S3. Based on the current detection data and voltage detection data of the target line, analyze the degree of current mutation and harmonic distortion of the target line, as well as the degree of change in the high-level voltage duty cycle and amplitude, and obtain the current transient anomaly index and voltage transient anomaly index through linear weighted fusion.

[0012] S4. Substitute the current transient abnormality index and voltage transient abnormality index of the target line into the preset logistic regression formula to calculate the probability of a line contact fault. If the probability is greater than the preset probability threshold, it is determined that a line contact fault has occurred in the target line.

[0013] S5. Continue to track the current and voltage transient abnormality indicators of the target line in multiple subsequent sampling time intervals, and determine whether the target line contact fault is a permanent or transient contact fault through statistical testing methods.

[0014] S6. Issue a line contact fault warning to the target line and simultaneously send a line contact fault type identification code.

[0015] Compared with the prior art, the present invention has the following beneficial effects: (1) By comparing with historical benchmark values, the present invention dynamically adjusts the sampling frequency within a preset sampling time interval if the current or voltage of the target line exceeds the limit. By dynamically adjusting the sampling frequency, the sampling frequency can be quickly increased when abnormal changes occur in the current or voltage at the early stage of a line fault, thereby accurately capturing key transient information at the early stage of the fault.

[0016] (2) Based on the current detection data and voltage detection data of the target line, the present invention not only analyzes the degree of current mutation and harmonic distortion of the target line, but also pays attention to the high-level voltage duty cycle and amplitude change degree, comprehensively evaluates the electrical status of the target line from multiple angles, accurately reflects the current and voltage transient abnormality indicators of the line-touching fault, and comprehensively analyzes the current, voltage transient abnormality indicators and environmental assessment factors to calculate the probability of line-touching fault, taking into account the influence of environmental parameters, thereby improving the accuracy of fault identification and avoiding misjudgment or missed judgment.

[0017] (3) The present invention classifies line contact fault types into transient line contact faults and permanent line contact faults based on the duration of the line contact faults. Continuous monitoring and statistical testing are carried out on the transient abnormality indicators of current and voltage within each sampling time interval. The line contact faults of the target line are classified according to the results of the statistical testing, thereby achieving a quantitative assessment of the urgency of the line contact faults and improving the response efficiency of the line contact fault handling process. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 The present invention is a flowchart of the steps for implementing the method.

[0020] Figure 2 This is a logic flow chart for dynamically adjusting the sampling frequency within the detection sampling time interval in step S2 of the present invention.

[0021] Figure 3 This is a logic flow chart for obtaining the degree of current harmonic distortion in step S3 of the present invention. DETAILED DESCRIPTION

[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present invention 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 making creative efforts are within the scope of protection of the present invention.

[0023] See also Figure 1 As shown, the present invention provides a line contact fault identification method based on AC / DC system, including: S1. selecting a single line between adjacent detection points as the target line, and collecting current data and voltage data of the target line at a preset sampling frequency.

[0024] It should be noted that, given that the maximum frequency of the power system signal is 2kHz, according to the Nyquist sampling theorem, the sampling frequency should not be less than twice the maximum frequency of the signal. However, in actual engineering applications, the signal bandwidth will be limited to below 1kHz, so the above-mentioned preset sampling frequency can be set to 2kHz for example to accurately reconstruct the power system signal.

[0025] S2. Compare with historical benchmark values. If the current or voltage of the target line exceeds the limit, dynamically adjust the sampling frequency within the detection sampling time interval, and synchronously collect the current data and voltage data after the sampling frequency adjustment as the current detection data and voltage detection data of the target line.

[0026] As a preferred embodiment, the target line current historical reference value is specifically obtained by the following method: retrieving the historical normal operation data of the target line to determine the rated value of the normal current, calculating the product of the rated value of the normal current and a preset quantile, and using the product as the target line current historical reference value.

[0027] It should be noted that the above preset quantiles can be exemplarily set to .

[0028] It should be noted that the above-mentioned retrieval of the historical normal operating data of the target line to determine the rated value of the normal current includes: retrieving the historical operating current data of the target line for nearly one year, screening out the time period when the target line is in normal operating state through statistical analysis of the historical operating current data and combining it with the fault maintenance record of the target line, obtaining the current data within the above-mentioned normal operating time period, and using the median of the historical normal operating current data as the rated value of the normal current of the target line.

[0029] It should be noted that the target line voltage historical reference value is obtained in the same manner as the target line current historical reference value is obtained.

[0030] See also Figure 2 As shown, as a preferred embodiment, the sampling frequency within the detection sampling time interval is dynamically adjusted, and the specific acquisition method is as follows: the time point when the current current or voltage of the target line exceeds the limit is recorded as the adjustment trigger point, and a detection sampling time interval with the adjustment trigger point as the starting point is constructed according to the preset duration.

[0031] Calculate the amplitude deviation of the monitored current or voltage data from its historical reference value.

[0032] Based on the preset line sampling frequencies corresponding to the amplitude deviation intervals stored in the WEB cloud, the adjustment value of the sampling frequency within the detection sampling time interval is determined.

[0033] The embodiment of the present invention compares historical benchmark values. If the current or voltage of the target line exceeds the limit, the sampling frequency within the preset sampling time interval is dynamically adjusted. By dynamically adjusting the sampling frequency, when abnormal changes in current or voltage occur in the early stage of a line-touching fault, the sampling frequency can be quickly increased, thereby accurately capturing key transient information in the early stage of the fault.

[0034] S3. Based on the current detection data and voltage detection data of the target line, analyze the degree of current mutation and harmonic distortion of the target line, as well as the degree of change in the high-level voltage duty cycle and amplitude, and obtain the current transient anomaly index and voltage transient anomaly index through linear weighted fusion.

[0035] It should be noted that the above is to obtain the current transient anomaly index by linearly weighted fusion of the target line current mutation degree and the harmonic distortion degree, and to obtain the voltage transient anomaly index by linearly weighted fusion of the voltage high-level duty cycle and the amplitude change degree. The linear weight distribution can be set according to industry experience. For example, the weights of the current mutation degree and the harmonic distortion degree are set to 0.7 and 0.3 respectively. The reason is that the current mutation degree directly reflects the severity of the transient process, is more sensitive to line fault detection, and has a larger weight. The harmonic distortion degree reflects the quality of the current waveform, but is affected by the long-term operating state, and its contribution to transient anomalies is relatively small, so it has a smaller weight.

[0036] The weights of the high-level voltage duty cycle and amplitude change degree are set to 0.6 and 0.4, respectively. The reason is that the high-level voltage duty cycle directly reflects the time characteristics of voltage fluctuations and can intuitively reflect the voltage changes when a line fault occurs. It has a larger weight. The amplitude change degree reflects the instantaneous fluctuation of voltage, but is easily affected by noise or normal load changes, so it has a smaller weight.

[0037] Linear weight distribution can also be obtained through a limited number of test data. Taking the current transient abnormality index as an example, first collect historical data on the current transient abnormality index, current mutation degree and harmonic distortion degree, then calculate the correlation coefficient of the influence of different current mutation degrees and different harmonic distortion degrees on current transient abnormality, use regression analysis or logistic regression analysis to determine the contribution of current mutation degree and harmonic distortion degree, and finally, after normalization, convert the contribution into weight and its sum is 1.

[0038] As a preferred embodiment, the degree of current mutation is specifically obtained by the following method: calculating the current average value based on the current detection data of the target line, and calculating the normal current average value based on the historical normal operation data of the target line, subtracting the current average value from the normal current average value to obtain a current deviation value, and performing a ratio operation on the current deviation value and the normal current average value to obtain the overall current mutation rate.

[0039] According to the current detection data of the target line, the time domain waveform of the current of the target line is drawn, and the current values ​​corresponding to each peak and each trough in the waveform and the time coordinates of their occurrence are recorded. The absolute difference in current between each peak in the time domain waveform and its nearest preceding trough is calculated respectively. The absolute difference in current is then ratioed with the time interval between the two to obtain the current mutation rate between each peak in the time domain waveform and its nearest preceding trough.

[0040] The average, median and mode of the current mutation rate between each peak and its nearest preceding trough are calculated and used as reference values ​​in turn. The absolute cumulative error between the current mutation rate of each peak and its nearest preceding trough and each reference value is obtained, and the current mutation rate corresponding to the minimum absolute cumulative error is selected as the local current mutation rate.

[0041] Based on the overall current mutation rate and the local current mutation rate, the geometric mean calculation method is used to calculate the comprehensive current mutation rate, which is used as the current mutation degree.

[0042] See also Figure 3 As shown, as a preferred embodiment, the current harmonic distortion degree is specifically obtained by the following method: the current detection data of the target line is converted into the frequency domain by Fourier transform, and the fundamental current and each harmonic current component are separated.

[0043] Calculate the effective value of the fundamental current and each harmonic current.

[0044] The Euclidean norm of each harmonic current effective value is calculated based on the Euclidean norm formula to serve as the total harmonic current effective value.

[0045] The current harmonic distortion rate is obtained by performing a ratio operation on the total harmonic current effective value and the fundamental current effective value, and is used as the current harmonic distortion degree of the target line.

[0046] It should be noted that the above-mentioned process of obtaining the separated fundamental current and each harmonic current component includes: performing Fourier transform on the current detection data of the target line to convert it into a frequency domain signal, finding the frequency component with the largest amplitude in the frequency domain signal and using it as the fundamental frequency, locating the current component with the same frequency in the frequency domain according to the fundamental frequency, using it as the fundamental current, and according to the fact that the frequency of each harmonic current is an integer multiple of the fundamental frequency, counting the components corresponding to the same frequency in the spectrum to obtain each harmonic current component.

[0047] It should be noted that the above-mentioned calculation of the effective value of the fundamental current and each harmonic current includes the following specific processes: extracting the amplitude of the fundamental component and the amplitude of each harmonic component from the frequency domain signal, multiplying the amplitude of the fundamental component and the amplitude of each harmonic component by a preset amplitude correction coefficient respectively to obtain the corrected amplitude of the fundamental component and the amplitude of each harmonic component, and performing ratio operations on the corrected amplitude of the fundamental component and the amplitude of each harmonic component with the total number of samples of the target line current detection data in turn to obtain the effective value of the fundamental current and the effective value of each harmonic current.

[0048] It should be noted that the above-mentioned preset amplitude correction coefficient can be set to 2 for example.

[0049] As a preferred embodiment, the voltage high-level duty cycle is specifically obtained by the following method: based on the voltage detection data of the target line, the time interval from the voltage data exceeding the voltage historical reference value to the voltage returning to the voltage historical reference value is taken as the high-level duration period, thereby obtaining each high-level duration period in the detection sampling time interval, and counting the total high-level duration in the detection sampling time interval, and performing a ratio analysis with the total duration of the detection sampling time interval to obtain the voltage high-level duty cycle.

[0050] As a preferred embodiment, the voltage amplitude variation degree is specifically obtained by integrating the difference between the voltage data in each high-level duration period and the voltage historical reference value to obtain the voltage cumulative deviation in each high-level duration period.

[0051] The product of the total duration of the high level and the voltage historical reference value is used as the voltage standard reference, and the voltage cumulative deviation in each high level duration period is ratioed to the voltage standard reference, and the maximum value is selected as the voltage comprehensive amplitude change rate, which is used as the voltage amplitude change degree.

[0052] The embodiment of the present invention not only analyzes the degree of current mutation and harmonic distortion of the target line based on the current detection data and voltage detection data of the target line, but also pays attention to the high-level voltage duty cycle and amplitude change degree, comprehensively evaluates the electrical status of the target line from multiple angles, accurately reflects the current and voltage transient abnormality indicators of the line-touching fault, and takes into account the influence of environmental parameters when further analyzing the current and voltage transient abnormality indicators to calculate the probability of the line-touching fault, thereby improving the accuracy of fault identification and avoiding misjudgment or missed judgment.

[0053] S4. Substitute the current transient abnormality index and voltage transient abnormality index of the target line into the preset logistic regression formula to calculate the probability of a line contact fault. If the probability is greater than the preset probability threshold, it is determined that a line contact fault has occurred in the target line.

[0054] As a preferred method, the current transient abnormality index and voltage transient abnormality index of the target line are substituted into a preset logistic regression formula to calculate the probability of line contact fault. The specific acquisition method is as follows: From the formula The probability of line-touching fault P is calculated, where 、 is the preset regression coefficient, is the preset intercept, 、 They are the current transient abnormality index and voltage transient abnormality index in the detection sampling time interval, is the environmental assessment factor, 、 The electrical assessment weight and environmental assessment weight of the preset line-touching fault are respectively is a preset natural constant.

[0055] It should be noted that, given that the value range of the probability of a line-touching fault is [0, 1], the above-mentioned preset probability threshold can be exemplarily set to 0.7.

[0056] It should be noted that the specific reasons for selecting the preset logistic regression formula to calculate the probability of line-touching faults are as follows: there is a nonlinear relationship between the probability of line-touching faults and transient abnormality indicators. The preset logistic regression formula can effectively capture the above nonlinear relationship and improve the accuracy of judgment. Its smoothing characteristics can suppress the influence of abnormal data and have strong anti-noise ability. In addition, the formula can directly output the probability of fault occurrence, which is convenient for application in engineering.

[0057] It should be noted that the electrical assessment weight and environmental assessment weight of the above-mentioned preset line-touching fault can be set as 0.8 and 0.2, respectively. The reason for this is that when a line-touching fault occurs, the electrical parameters of the target line change significantly, so that the abnormal changes in the electrical parameters have a dominant influence on the probability of a line-touching fault. Therefore, the electrical assessment has a higher weight, and the data collected by the environmental assessment is easily affected by interference factors such as meteorological conditions, and cannot accurately assess the probability of a line-touching fault, and therefore has a smaller weight.

[0058] It should be noted that the above-mentioned preset regression coefficients and preset intercepts are specifically obtained as follows: retrieve the historical current and voltage operating data of the target line in the past two years, and filter out the historical line-touching fault data based on the fault maintenance records of the target line, calculate the current transient abnormality index and voltage transient abnormality index corresponding to each historical line-touching fault data, retrieve the environmental data corresponding to each historical line-touching fault data and calculate the corresponding environmental assessment factor.

[0059] The frequency of line-touching faults under different current transient abnormality indicators, voltage transient abnormality indicators and environmental assessment factors is counted, and the frequency is normalized to the probability of line-touching faults.

[0060] The initial values ​​of the regression coefficient and the intercept are set. Based on the preset logistic regression formula, the current transient abnormal characteristic value, the voltage transient abnormal characteristic value, the environmental impact assessment factor and the normalized probability of occurrence of the line contact fault corresponding to the historical line contact fault data are substituted into professional fitting processing software, such as the fitting toolbox of Matlab, and the preset regression coefficient and the preset intercept are obtained by solving the fitting toolbox using the fminunc optimization algorithm.

[0061] As a preferred embodiment, the environmental assessment factor is specifically obtained by the following method: collecting a real-time environmental data set of the target line, including noise, temperature and smoke concentration data, retrieving a historically normal operating environmental data set of the target line, selecting the maximum values ​​of the normal operating data of noise, temperature and smoke concentration and performing normalization processing, and using the maximum values ​​of the normal operating data of noise, temperature and smoke concentration after the normalization processing as the environmental standard data set.

[0062] The noise, temperature and smoke concentration data of the real-time environmental data set are standardized, and the difference operation is performed between the noise, temperature and smoke concentration data of the standardized real-time environmental data set and the corresponding environmental parameters in the environmental standard data set to obtain the deviation value of each environmental parameter. If the deviation value of an environmental parameter is less than 0, the risk indicator value of the environmental parameter is set to 0. Otherwise, the deviation value of the environmental parameter is synchronized as the risk indicator value. In this way, the risk indicator values ​​of each environmental parameter are accumulated to obtain the environmental assessment factor.

[0063] S5. Continue to track the current and voltage transient abnormality indicators of the target line in multiple subsequent sampling time intervals, and determine whether the target line contact fault is a permanent or transient contact fault through statistical testing methods.

[0064] As a preferred method, the target line contact fault is determined to be permanent or transient by the statistical test method. The specific acquisition method is as follows: calculate the average value and variance of the current transient abnormality index of the target line in each subsequent sampling time interval, and calculate the average value and variance of the current transient abnormality index of the target line according to the formula Calculate the difference between the subsequent sampling time interval and the detection sampling time interval relative to the current transient abnormality index, where 、 are the average value and variance of the current transient abnormality index of the target line in each sampling time interval, is the number of sampling time intervals.

[0065] Similarly, the difference between the subsequent sampling time interval and the detection sampling time interval relative to the voltage transient anomaly index can be obtained.

[0066] The preset difference thresholds corresponding to the current transient anomaly index and voltage transient anomaly index stored in the WEB cloud are extracted. If the differences between the current transient anomaly index and the voltage transient anomaly index in the subsequent sampling time interval and the detection sampling time interval are both greater than the corresponding difference thresholds, the target line fault is determined to be a transient fault; otherwise, the target line fault is determined to be a permanent fault.

[0067] It should be noted that the above-mentioned preset difference threshold is specifically obtained in the following manner: by retrieving the fault maintenance records of the target line to filter out the instantaneous line-touching fault data, the corresponding current transient abnormality index and voltage transient abnormality index are calculated, and for the filtered instantaneous line-touching fault data, the mean and standard deviation of the current transient abnormality index and the voltage transient abnormality index are calculated respectively, and the mean plus three times the standard deviation is used as the difference threshold corresponding to the current transient abnormality index and the voltage transient abnormality index.

[0068] The embodiment of the present invention divides the line contact fault types into transient line contact faults and permanent line contact faults according to the duration of the line contact faults, conducts continuous monitoring and statistical testing on the transient abnormal indicators of current and voltage in each sampling time interval, and classifies the line contact faults of the target line according to the results of the statistical test, thereby achieving a quantitative assessment of the urgency of the line contact faults and improving the response efficiency of the line contact fault handling process.

[0069] S6. Issue a line contact fault warning to the target line and simultaneously send a line contact fault type identification code.

[0070] The above formulas are all dimensionless and numerically calculated, and the preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0071] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0072] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0073] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0074] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0075] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying line contact faults based on AC / DC systems, characterized in that: include: S1. Select a single line between adjacent detection points as the target line and collect current and voltage data of the target line at a preset sampling frequency; S2. Compare historical benchmark values. If the current or voltage of the target line exceeds the limit, dynamically adjust the sampling frequency within the detection sampling time interval and collect the current detection data and voltage detection data of the target line after the adjustment; S3. Analyze the target line current mutation degree, harmonic distortion degree, high voltage duty cycle, and amplitude change degree based on the target line current and voltage data, and obtain the current transient anomaly index and voltage transient anomaly index through linear weighted fusion; S4. Calculate the probability of a line-touching fault based on the current transient abnormality index and the voltage transient abnormality index using a logistic regression algorithm. If the probability is greater than a preset probability threshold, determine that a line-touching fault has occurred on the target line; S5. Track the current and voltage transient abnormality indicators of the target line in multiple subsequent sampling time intervals, and determine whether the target line contact fault is permanent or transient through statistical testing methods; S6. Issue a line fault warning for the target line and simultaneously send a line fault type identification code; Substitute the current transient abnormality index and voltage transient abnormality index of the target line into the preset logistic regression formula to calculate the probability of line contact fault. The specific acquisition method is as follows: By the formula The probability of line-touching fault P is calculated, where 、 is the preset regression coefficient, is the preset intercept, 、 They are the current transient abnormality index and voltage transient abnormality index in the detection sampling time interval, is the environmental assessment factor, 、 The electrical assessment weight and environmental assessment weight of the preset line-touching fault are respectively To presuppose a natural constant; The target line contact fault is determined to be permanent or transient by statistical testing. The specific acquisition method is as follows: Calculate the average value and variance of the current transient abnormality index of the target line in each subsequent sampling time interval, according to the formula Calculate the difference between the subsequent sampling time interval and the detection sampling time interval relative to the current transient abnormality index, where 、 are the average value and variance of the current transient abnormality index of the target line in each sampling time interval, is the number of sampling time intervals; Similarly, the difference between the subsequent sampling time interval and the detection sampling time interval relative to the voltage transient abnormality index can be obtained; The preset difference thresholds corresponding to the current transient anomaly index and voltage transient anomaly index stored in the WEB cloud are extracted. If the differences between the current transient anomaly index and the voltage transient anomaly index in the subsequent sampling time interval and the detection sampling time interval are both greater than the corresponding difference thresholds, the target line fault is determined to be a transient fault; otherwise, the target line fault is determined to be a permanent fault.

2. The method for identifying line contact faults based on AC / DC systems according to claim 1, characterized in that: The target line current historical reference value is obtained in the following specific method: The historical normal operation data of the target line is retrieved to determine the rated value of the normal current, the product of the rated value of the normal current and a preset quantile is calculated, and the product is used as the historical reference value of the target line current.

3. The method for identifying line contact faults based on AC / DC systems according to claim 1, characterized in that: The specific acquisition method of the dynamic adjustment detection sampling frequency within the sampling time interval is as follows: The time point when the current or voltage of the target line exceeds the limit is recorded as the adjustment trigger point, and a detection sampling time interval with the adjustment trigger point as the starting point is constructed according to the preset duration; Calculate the amplitude deviation between the monitored current or voltage data and its historical reference value; Based on the preset line sampling frequencies corresponding to the amplitude deviation intervals stored in the WEB cloud, the adjustment value of the sampling frequency within the detection sampling time interval is determined.

4. The method for identifying line contact faults based on AC / DC systems according to claim 1, characterized in that: The specific method for obtaining the degree of current mutation is as follows: Calculating a current average value based on the current detection data of the target line, and calculating a normal current average value based on the historical normal operation data of the target line, subtracting the current average value from the normal current average value to obtain a current deviation value, and performing a ratio operation on the current deviation value and the normal current average value to obtain an overall current mutation rate; Based on the current detection data of the target line, the time domain waveform of the target line current is plotted, and the current values ​​corresponding to each peak and each trough in the waveform and the time coordinates of their occurrence are recorded. The absolute difference between the current of each peak and its nearest preceding trough in the time domain waveform is calculated, and the absolute difference between the current and the time interval between the two is calculated to obtain the current mutation rate between each peak and its nearest preceding trough in the time domain waveform; Calculate the average, median, and mode of the current mutation rate between each peak and its nearest preceding trough and use them as reference values ​​in turn. Obtain the absolute cumulative error between the current mutation rate of each peak and its nearest preceding trough and each reference value. Filter the current mutation rate corresponding to the minimum absolute cumulative error as the local current mutation rate. Based on the overall current mutation rate and the local current mutation rate, the geometric mean calculation method is used to calculate the comprehensive current mutation rate, which is used as the current mutation degree.

5. The method for identifying line contact faults based on AC / DC systems according to claim 1, characterized in that: The specific method for obtaining the current harmonic distortion degree is as follows: The current detection data of the target line is converted into the frequency domain through Fourier transform, and the fundamental current and each harmonic current component are separated; Calculate the effective value of fundamental current and each harmonic current; Calculate the Euclidean norm of each harmonic current effective value based on the Euclidean norm formula to obtain the total harmonic current effective value; The current harmonic distortion rate is obtained by performing a ratio operation on the total harmonic current effective value and the fundamental current effective value, and is used as the current harmonic distortion degree of the target line.

6. The method for identifying line contact faults based on AC / DC systems according to claim 1, characterized in that: The specific method for obtaining the voltage high level duty cycle is as follows: According to the voltage detection data of the target line, the time interval from when the voltage data exceeds the voltage historical reference value to when it recovers to the voltage historical reference value is taken as the high-level duration period, so as to obtain each high-level duration period in the detection sampling time interval, and the total high-level duration in the detection sampling time interval is counted, and the ratio analysis is performed with the total duration of the detection sampling time interval to obtain the voltage high-level duty cycle.

7. The method for identifying line contact faults based on AC / DC systems according to claim 6, characterized in that: The specific method for obtaining the voltage amplitude change degree is as follows: Integrate the difference between the voltage data in each high-level duration period and the voltage historical reference value to obtain the voltage cumulative deviation in each high-level duration period; The product of the total duration of the high level and the voltage historical reference value is used as the voltage standard reference, and the voltage cumulative deviation in each high level duration period is ratioed to the voltage standard reference, and the maximum value is selected as the voltage comprehensive amplitude change rate, which is used as the voltage amplitude change degree.

8. The method for identifying line contact faults based on AC / DC systems according to claim 1, characterized in that: The specific method for obtaining the environmental assessment factors is as follows: Collect the real-time environmental data set of the target line, including noise, temperature, and smoke concentration data. Retrieve the historical normal operation environmental data set of the target line. Select the maximum value of the normal operation data of noise, temperature, and smoke concentration and perform normalization processing. The maximum value of the normal operation data of noise, temperature, and smoke concentration after normalization processing is used as the environmental standard data set. The noise, temperature and smoke concentration data of the real-time environmental data set are standardized, and the difference operation is performed between the noise, temperature and smoke concentration data of the standardized real-time environmental data set and the corresponding environmental parameters in the environmental standard data set to obtain the deviation value of each environmental parameter. If the deviation value of an environmental parameter is less than 0, the risk indicator value of the environmental parameter is set to 0. Otherwise, the deviation value of the environmental parameter is synchronized as the risk indicator value. In this way, the risk indicator values ​​of each environmental parameter are accumulated to obtain the environmental assessment factor.

Images