Cable partial discharge detection method and system under high noise

Abstract

The invention relates to the technical field of power equipment detection, and particularly discloses a cable partial discharge detection method and system under high noise, and the method comprises the steps: dividing a cable into a plurality of monitoring regions, employing a current sensor to collect a current pulse signal of each region in real time, and enabling the current pulse signal to comprise the current pulse frequency and repetition rate; current pulse frequency fluctuation characteristics and current pulse repetition rate characteristics are extracted, and corresponding characteristic indexes are generated; then, by comprehensively analyzing the characteristic index of each region, calculating a comprehensive index and comparing the comprehensive index with a preset threshold value, whether the partial discharge phenomenon exists in the region or not is judged; further analyzing the deviation between the real-time discharge frequency and the historical normal discharge data of the region which is judged to be partial discharge, performing anomaly detection by using an isolated forest model, and judging whether the discharge is abnormal or not; when abnormal discharge is detected, the system automatically starts an early warning and regulation and control mechanism to carry out voltage regulation or temperature control so as to avoid equipment damage or faults.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment detection, and in particular to a cable partial discharge detection method and system under high noise. Background Art

[0002] In the operation of modern power systems, communication networks, and high-precision equipment, cables, as important connecting elements, carry the functions of power transmission and data transmission. However, during long-term use, cables may be affected by factors such as the external environment, aging, and mechanical damage, leading to the occurrence of partial discharge. Partial discharge will not only reduce the electrical performance of the cable, but may also cause problems such as aging of the insulation material and insulation breakdown. In severe cases, it may cause equipment failure or safety accidents. Therefore, timely detection and diagnosis of partial discharge is crucial to ensure the safe and stable operation of power equipment and communication facilities.

[0003] The detection methods of partial discharge mainly include traditional electrical detection method, optical detection method, ultrasonic detection method, etc. Although these methods can detect partial discharge to a certain extent, they face some problems in practical application. The existing technology still has problems such as signal interference in high noise environment, insufficient detection accuracy and low automation level.

[0004] First, in a high-noise environment, these traditional methods are susceptible to external interference, resulting in false alarms or missed alarms. Second, traditional detection methods usually rely on manual analysis and human judgment, lack sufficient real-time and intelligence, and are difficult to monitor and analyze the discharge situation comprehensively and accurately. In addition, the spatial resolution of traditional methods is usually low, making it difficult to effectively locate and judge the specific location and status of the discharge, especially over long distances or in complex environments. Summary of the invention

[0005] The object of the present invention is to provide a method and system for detecting partial discharge of cables under high noise conditions, so as to solve the above-mentioned problems.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A method for detecting partial discharge of a cable under high noise conditions comprises the following steps:

[0008] S1: During the monitoring period, the cable is divided into several identical areas, and the current pulse signal in each area is obtained respectively;

[0009] Wherein, the current pulse signal is collected by a current sensor, including the current pulse frequency and the current pulse repetition rate;

[0010] S2: preprocessing the current pulse signal in each region, extracting the current pulse frequency fluctuation characteristics and the current pulse repetition rate characteristics, and generating the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index;

[0011] S3: Comprehensively analyzing the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each area, calculating the comprehensive index of each area, and determining whether the corresponding area is a partial discharge area;

[0012] S4: comparing and analyzing the real-time discharge frequency and discharge repetition rate of the determined partial discharge area with the historical normal discharge frequency and discharge repetition rate to determine whether the partial discharge is abnormal;

[0013] S5: According to the judgment result, when partial discharge abnormality is detected, the system starts the early warning adjustment mechanism.

[0014] As a further solution of the present invention: the process of obtaining the current pulse frequency fluctuation characteristic index specifically includes:

[0015] Obtaining current pulse frequency data in each area;

[0016] Preprocessing the current pulse frequency data;

[0017] Identify all local maxima and local minima in the current pulse frequency data;

[0018] Use spline interpolation to connect local maximum and minimum values ​​to generate upper and lower envelopes;

[0019] Calculate the mean of the upper and lower envelopes;

[0020] The first intrinsic mode function is extracted from the current pulse frequency data using the empirical mode decomposition method;

[0021] Repeat the steps for the remaining signals to continue extracting all the intrinsic mode functions;

[0022] Perform Hilbert transform on each eigenmode function to obtain the analytical signal;

[0023] Calculate the instantaneous amplitude and instantaneous phase of the analytical signal;

[0024] The instantaneous frequency is calculated by taking the derivative of the instantaneous phase with respect to time, and the standard deviation σ of the instantaneous frequency is calculated. f and the average value μ f , the current pulse frequency fluctuation characteristic index is calculated based on the standard deviation and average value of the instantaneous frequency.

[0025] As a further solution of the present invention: the process of obtaining the current pulse repetition rate characteristic index specifically includes:

[0026] Acquire the current pulse repetition rate data in each region to form a data matrix X;

[0027] The current pulse repetition rate data is standardized to obtain the standardized data matrix Z.

[0028] According to the standardized data matrix Z, calculate the covariance matrix C;

[0029] Perform eigenvalue decomposition on the covariance matrix C to obtain eigenvalues ​​and eigenvectors;

[0030] Select the first k largest eigenvalues ​​and their corresponding eigenvectors to form the principal component matrix V k ;

[0031] Where k represents the number of eigenvalues;

[0032] Project the standardized data into the principal component space to obtain the reduced-dimensional data Z k ;

[0033] Reconstruct the original data according to the principal component projection results and obtain the reconstructed data matrix Z s ;

[0034] Calculate the reconstruction error matrix E;

[0035] Calculate the reconstruction error norm RE for each data point i ;

[0036] Among them, the calculation expression of the reconstruction error norm is:

[0037] RE i =||E i ||;

[0038] In the formula, i represents the number of data points, RE i represents the reconstruction error norm of the i-th data point, E i represents the reconstruction error matrix of the i-th data point;

[0039] Calculate the mean μ of the reconstruction error norm F and standard deviation σ F , and the mean and standard deviation of the reconstruction error norm are normalized to calculate the current pulse repetition rate characteristic index.

[0040] As a further solution of the present invention: the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each region are comprehensively analyzed to calculate the comprehensive index of each region, specifically including:

[0041] The current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each region are obtained, and the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each region are normalized to calculate the comprehensive index of each region.

[0042] As a further solution of the present invention: the step of determining whether the corresponding area is a partial discharge area specifically includes:

[0043] It is determined whether the comprehensive index is less than a preset comprehensive index threshold. If so, it is recorded as a partial discharge area; if not, it is recorded as a non-partial discharge area.

[0044] As a further solution of the present invention: comparing and analyzing the real-time discharge frequency and discharge repetition rate of the determined partial discharge area with the historical normal discharge frequency and discharge repetition rate to determine whether the partial discharge is abnormal, specifically includes:

[0045] Obtain the historical normal discharge frequency and discharge repetition rate of multiple collection points according to the time series, and record them as historical normal data;

[0046] Standardize historical data;

[0047] The processed historical data is used as a training set to train the isolation forest model;

[0048] The average isolation degree of the corresponding historical data points is calculated through the isolation depth of multiple trees;

[0049] Isolation Forest partitions historical data using multiple trees and ultimately generates an isolation score for each historical data point;

[0050] The calculation expression of the isolation score is:

[0051]

[0052] Where h(x) represents the isolation score of historical data point x, H(x) represents the average isolation depth in the isolation tree of historical data point x, m represents the number of historical data in the data set, and c(m) represents the maximum isolation depth in the isolation tree;

[0053] The normal discharge frequency and discharge repetition rate of each area are collected in real time according to the time series and recorded as regional real-time data;

[0054] Standardize real-time data;

[0055] The processed real-time data of each region is input into the trained isolation forest model for detection;

[0056] The isolation score of each regional real-time data point is calculated using the isolation forest model;

[0057] For each region, the average isolation score of all real-time data points is calculated and recorded as the real-time region average isolation score;

[0058] For each region, the average isolation score of all historical data points is calculated and recorded as the historical region average isolation score;

[0059] The difference between the real-time regional average isolation score and the historical regional average isolation score is calculated, and the absolute value is removed and recorded as the regional deviation value;

[0060] comparing the region deviation value to a region deviation value threshold;

[0061] If the regional deviation value is greater than or equal to the regional deviation value threshold, it is recorded as an abnormal area;

[0062] If the area deviation value is less than the area deviation value threshold, it is recorded as a normal area.

[0063] As a further solution of the present invention: according to the judgment result, when the partial discharge abnormality is detected, the system starts the early warning adjustment mechanism, which specifically includes:

[0064] Based on the abnormal area, the system will automatically take alarm response;

[0065] After the alarm, the system automatically takes control measures, including:

[0066] Calculate the standard deviation and average value of the voltage in the corresponding area in the period before the warning, calculate the ratio of the standard deviation to the average value of the voltage, and obtain the voltage jump ratio;

[0067] comparing the voltage jump ratio with a preset threshold;

[0068] If the voltage jump ratio is greater than or equal to the preset threshold, the real-time voltage is adjusted according to the historical normal voltage range;

[0069] If the voltage jump ratio is less than the preset threshold, the temperature of the insulation layer is detected to determine whether the temperature of the insulation layer is greater than or equal to the preset temperature threshold. If so, the temperature is lowered through the temperature control system;

[0070] After adjusting the voltage and temperature of the cable during operation, re-judge the discharge situation in the corresponding area. If abnormal discharge still exists in the area, power-off isolation measures are taken.

[0071] A cable partial discharge detection system under high noise, comprising:

[0072] A data acquisition module, wherein the data acquisition module is used to divide the cable into a number of identical areas during a monitoring period and obtain a current pulse signal in each area;

[0073] A data preprocessing module, which preprocesses the current pulse signal in each area, extracts the current pulse frequency fluctuation characteristics and the current pulse repetition rate characteristics, and generates a current pulse frequency fluctuation characteristic index and a current pulse repetition rate characteristic index;

[0074] A partial discharge area determination module, wherein the partial discharge area determination module calculates a comprehensive index of each area by comprehensively analyzing a current pulse frequency fluctuation characteristic index and a current pulse repetition rate characteristic index in each area, and determines whether the corresponding area is a partial discharge area;

[0075] A partial discharge abnormality analysis module, which is used to compare and analyze the real-time discharge frequency and discharge repetition rate of the determined partial discharge area with the historical normal discharge frequency and discharge repetition rate to determine whether the partial discharge is abnormal;

[0076] The early warning adjustment module starts the early warning adjustment mechanism according to the judgment result when the partial discharge abnormality is detected.

[0077] A cable partial discharge detection system under high noise, the data preprocessing module at least further comprises:

[0078] a current pulse frequency fluctuation characteristic index calculation unit, wherein the current pulse frequency fluctuation characteristic index calculation unit calculates the current pulse frequency fluctuation characteristic index by processing the current pulse frequency;

[0079] A current pulse repetition rate characteristic index calculation unit is used to calculate the current pulse repetition rate characteristic index by processing the current pulse repetition.

[0080] A cable partial discharge detection system under high noise, wherein the partial discharge area determination module at least includes:

[0081] A comprehensive index calculation unit is provided, wherein the comprehensive index is obtained by comprehensively calculating the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index.

[0082] Beneficial effects of the present invention:

[0083] (1) By dividing the cable into regions and installing current sensors in each region, partial discharge signals can be accurately captured. This method combines real-time monitoring of current pulse frequency and repetition rate, and through advanced signal preprocessing technology, can effectively remove noise interference and extract accurate discharge characteristics. This method enables cable partial discharge detection to have a higher spatial resolution, and can promptly detect early discharge anomalies, thereby effectively avoiding equipment damage or safety hazards. Compared with traditional methods, this method can provide more accurate discharge detection in a high-noise environment, avoiding the problem of misjudgment or missed judgment caused by noise in traditional methods, significantly improving the stability and monitoring accuracy of the system, and has strong practical application value.

[0084] (2) The present invention improves the intelligence and accuracy of discharge anomaly detection by introducing comprehensive index analysis and isolation forest model; when detecting partial discharge anomaly, the system can automatically start the early warning and control mechanism, and take timely emergency response measures for the abnormal area; this mechanism includes real-time adjustment of voltage and temperature, and can judge whether there are abnormal fluctuations based on historical data, so as to take corresponding control measures to avoid partial discharge from causing long-term damage to the equipment. Through this intelligent adjustment mechanism, the system can not only improve the fault response speed, but also intervene before the equipment becomes abnormal, thereby improving the safety and reliability of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] The present invention will be further described below in conjunction with the accompanying drawings.

[0086] Figure 1 It is a flowchart of the specific steps of a cable partial discharge detection method under high noise of the present invention;

[0087] Figure 2 It is a flow chart of a cable partial discharge detection system under high noise in the present invention. DETAILED DESCRIPTION

[0088] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0089] See also Figure 1 As shown, the present invention is a method for detecting partial discharge of cables under high noise, comprising the following steps:

[0090] S1: During the monitoring period, the cable is divided into several identical areas, and the current pulse signal in each area is obtained respectively;

[0091] Wherein, the current pulse signal is collected by a current sensor, including the current pulse frequency and the current pulse repetition rate;

[0092] S2: preprocessing the current pulse signal in each region, extracting the current pulse frequency fluctuation characteristics and the current pulse repetition rate characteristics, and generating the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index;

[0093] S3: Comprehensively analyzing the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each area, calculating the comprehensive index of each area, and determining whether the corresponding area is a partial discharge area;

[0094] S4: comparing and analyzing the real-time discharge frequency and discharge repetition rate of the determined partial discharge area with the historical normal discharge frequency and discharge repetition rate to determine whether the partial discharge is abnormal;

[0095] S5: According to the judgment result, when partial discharge abnormality is detected, the system starts the early warning adjustment mechanism.

[0096] In S1: During the monitoring period, the cable is divided into several identical areas, and the current pulse signal in each area is obtained respectively, including:

[0097] During the monitoring period, the entire cable is divided into several monitoring areas of equal length;

[0098] It should be noted that the length of each monitoring area is selected based on parameters such as the total length of the cable, the monitoring accuracy requirements, and the sensitivity of the equipment, so as to ensure that each area can representatively reflect the distribution of partial discharge signals after division;

[0099] After the area is divided, the system installs current sensors on each area;

[0100] The current pulse signal is captured in real time by the current sensor, including the current pulse frequency and current pulse repetition rate collected in real time;

[0101] Before data collection, the current sensors need to be calibrated to the same sampling frequency to ensure the time synchronization of the collected data and the accuracy of the collected data;

[0102] It should be noted that during the acquisition process, the original current pulse signal captured by the sensor includes the discharge signal generated by the partial discharge and the environmental noise signal;

[0103] The system will automatically remove DC bias and low-frequency interference and initially separate the useful signal;

[0104] The data of each monitoring area independently reflects the local discharge situation;

[0105] After each monitoring cycle, the system automatically stores the current pulse frequency and current pulse repetition rate of each area.

[0106] In S2, the current pulse signal in each region is preprocessed to extract the current pulse frequency fluctuation characteristics and the current pulse repetition rate characteristics, and generate the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index, which specifically include:

[0107] Obtaining current pulse frequency data in each area;

[0108] Preprocessing the current pulse frequency data, including filtering and denoising;

[0109] Identify all local maxima and local minima in the current pulse frequency data;

[0110] Use spline interpolation to connect local maximum and minimum values ​​to generate upper and lower envelopes;

[0111] Calculate the mean of the upper and lower envelopes;

[0112] The first intrinsic mode function is extracted from the current pulse frequency data using the empirical mode decomposition method;

[0113] Repeat the steps for the remaining signals to continue extracting all the intrinsic mode functions;

[0114] Perform Hilbert transform on each eigenmode function to obtain the analytical signal;

[0115] Calculate the instantaneous amplitude and instantaneous phase of the analytical signal;

[0116] The instantaneous frequency is calculated by taking the derivative of the instantaneous phase with respect to time, and the standard deviation σ of the instantaneous frequency is calculated. f and the average value μ f , calculate the current pulse frequency fluctuation characteristic index according to the standard deviation and average value of the instantaneous frequency;

[0117] Among them, the calculation expression of the current pulse frequency fluctuation characteristic index is:

[0118]

[0119] In the formula, I f Represents the current pulse frequency fluctuation characteristic index, σ f represents the standard deviation of the instantaneous frequency, μ f Represents the average value of the instantaneous frequency.

[0120] Acquire the current pulse repetition rate data in each region to form a data matrix X;

[0121] The current pulse repetition rate data is standardized to obtain the standardized data matrix Z.

[0122] According to the standardized data matrix Z, the covariance matrix C is calculated, and the calculation expression is:

[0123]

[0124] In the formula, n represents the number of data points, n is a positive integer greater than 1, C represents the covariance matrix, and Z T represents the transposed matrix of the data matrix Z;

[0125] Perform eigenvalue decomposition on the covariance matrix C to obtain eigenvalues ​​and eigenvectors;

[0126] Select the first k largest eigenvalues ​​and their corresponding eigenvectors to form the principal component matrix V k ;

[0127] Where k represents the number of eigenvalues;

[0128] Project the standardized data into the principal component space to obtain the reduced-dimensional data Z k ;

[0129] Reconstruct the original data according to the principal component projection results and obtain the reconstructed data matrix Z s , the calculation expression is:

[0130]

[0131] In the formula, Represents the principal component matrix V k The transposed matrix, Z s Represents the reconstructed data matrix, Z k Represents the data Z after dimensionality reduction k ;

[0132] Calculate the reconstruction error matrix E, the calculation expression is:

[0133] E=ZZ s ;

[0134] In the formula, E represents the reconstruction error matrix, and Z represents the standardized data matrix;

[0135] Calculate the reconstruction error norm RE for each data point i , the calculation expression is:

[0136] RE i =||E i ||;

[0137] In the formula, i represents the number of data points, RE irepresents the reconstruction error norm of the i-th data point, E i represents the reconstruction error matrix of the i-th data point;

[0138] Calculate the mean μ of the reconstruction error norm F and standard deviation σ F , and the mean and standard deviation of the reconstruction error norm are normalized to calculate the current pulse repetition rate characteristic index. The calculation expression is:

[0139]

[0140] In the formula, I F Represents the characteristic index of current pulse repetition rate, μ F represents the mean of the reconstruction error norm, σ F represents the standard deviation of the reconstruction error norm, and e represents the logarithm of the base of the natural number.

[0141] In S3, the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each region are comprehensively analyzed, and the comprehensive index of each region is calculated to determine whether the corresponding region is a partial discharge region, which specifically includes:

[0142] Obtaining a current pulse frequency fluctuation characteristic index and a current pulse repetition rate characteristic index in each region, normalizing the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each region, and calculating a comprehensive index of each region;

[0143] The calculation expression of the comprehensive index is:

[0144]

[0145] In the formula, j represents the number of regions, P j represents the comprehensive index in the jth region, I fj represents the current pulse frequency fluctuation characteristic index in the jth region, I Fj represents the characteristic index of the current pulse repetition rate in the jth region, a1 and a2 are preset proportional coefficients, and a1 and a2 are both greater than 0;

[0146] It should be noted that the comprehensive index reflects the discharge situation in each monitoring area, and the smaller the comprehensive index is, the more likely the corresponding area is to have partial discharge;

[0147] Compare the composite index in each region with a preset composite index threshold;

[0148] If the comprehensive index is less than the preset comprehensive index threshold, it means that the corresponding area is a partial discharge area;

[0149] If the comprehensive index is greater than or equal to the preset comprehensive index threshold, it indicates that the corresponding area is a non-partial discharge area.

[0150] In S4, the real-time discharge frequency and discharge repetition rate of the determined partial discharge area are compared and analyzed with the historical normal discharge frequency and discharge repetition rate to determine whether the partial discharge is abnormal, specifically including:

[0151] Obtain the historical normal discharge frequency and discharge repetition rate of multiple collection points according to time series (such as every minute), and record them as historical normal data;

[0152] Standardize historical data;

[0153] The processed historical data is used as a training set to train the isolation forest model, including:

[0154] By generating multiple random trees, each tree divides the historical data by randomly selecting features and split points;

[0155] Divide each historical data point and calculate the depth at which the corresponding historical data point is isolated in the tree;

[0156] The shallower the isolation depth of a historical data point, the easier it is to be isolated and may be an anomaly.

[0157] The average isolation degree of the corresponding historical data points is calculated through the isolation depth of multiple trees;

[0158] Isolation Forest partitions historical data using multiple trees and ultimately generates an isolation score for each historical data point;

[0159] The calculation expression of the isolation score is:

[0160]

[0161] Where h(x) represents the isolation score of historical data point x, H(x) represents the average isolation depth in the isolation tree of historical data point x, m represents the number of historical data in the data set, and c(m) represents the maximum isolation depth in the isolation tree;

[0162] The normal discharge frequency and discharge repetition rate of each area are collected in real time according to a time series (such as every minute), and recorded as regional real-time data;

[0163] Input the real-time data of each region into the trained isolation forest model for detection;

[0164] The isolation forest model calculates the isolation score of each regional real-time data point as the output of the model;

[0165] For each region, the average isolation score of all real-time data points is calculated and recorded as the real-time region average isolation score;

[0166] For each region, the average isolation score of all historical data points is calculated and recorded as the historical region average isolation score;

[0167] The difference between the real-time regional average isolation score and the historical regional average isolation score is calculated, and the absolute value is removed and recorded as the regional deviation value;

[0168] comparing the region deviation value to a region deviation value threshold;

[0169] If the regional deviation value is greater than or equal to the regional deviation value threshold, it means that the discharge situation of the corresponding area is abnormal and is recorded as an abnormal area;

[0170] If the regional deviation value is less than the regional deviation value threshold, it means that the discharge condition of the corresponding area is normal, and it is recorded as an abnormal area, and it is recorded as a normal area;

[0171] It should be noted that the regional deviation value reflects whether the discharge condition of the region is abnormal, and the lower the regional deviation value, the lower the degree of abnormal discharge in the corresponding region.

[0172] In S5, according to the judgment result, when partial discharge abnormality is detected, the system starts the early warning adjustment mechanism, which specifically includes:

[0173] Based on the abnormal area, the system will automatically take alarm response;

[0174] After the alarm, the system automatically takes control measures, including:

[0175] Calculate the standard deviation and average value of the voltage in the corresponding area for a period of time (such as one hour) before the warning, and calculate the ratio of the standard deviation to the average value of the voltage to obtain the voltage jump ratio;

[0176] comparing the voltage jump ratio with a preset threshold;

[0177] If the voltage jump ratio is greater than or equal to the preset threshold, it means that the voltage in the corresponding area is abnormal, and the real-time voltage is adjusted according to the historical normal voltage range;

[0178] If the voltage jump ratio is less than the preset threshold, it means that the voltage in the corresponding area is normal, then the temperature of the insulation layer is detected to determine whether the temperature of the insulation layer is greater than or equal to the preset temperature threshold. If so, the temperature is lowered through the temperature control system;

[0179] After adjusting the voltage and temperature of the cable during operation, re-judge the discharge situation in the corresponding area. If abnormal discharge still exists in the area, power-off isolation measures are taken.

[0180] See also Figure 2 As shown, a cable partial discharge detection system under high noise conditions includes:

[0181] A data acquisition module, wherein the data acquisition module is used to divide the cable into a number of identical areas during a monitoring period and obtain a current pulse signal in each area;

[0182] A data preprocessing module, which preprocesses the current pulse signal in each area, extracts the current pulse frequency fluctuation characteristics and the current pulse repetition rate characteristics, and generates a current pulse frequency fluctuation characteristic index and a current pulse repetition rate characteristic index;

[0183] A partial discharge area determination module, wherein the partial discharge area determination module calculates a comprehensive index of each area by comprehensively analyzing a current pulse frequency fluctuation characteristic index and a current pulse repetition rate characteristic index in each area, and determines whether the corresponding area is a partial discharge area;

[0184] A partial discharge abnormality analysis module, which is used to compare and analyze the real-time discharge frequency and discharge repetition rate of the determined partial discharge area with the historical normal discharge frequency and discharge repetition rate to determine whether the partial discharge is abnormal;

[0185] The early warning adjustment module starts the early warning adjustment mechanism according to the judgment result when the partial discharge abnormality is detected.

[0186] The working principle of the present invention is as follows: by dividing the cable into multiple monitoring areas, and collecting the current pulse signal of each area in real time through the current sensor, including the pulse frequency and repetition rate, to detect the partial discharge situation. First, in the preprocessing stage, the current pulse signal of each area is denoised and feature extracted to obtain the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index. Then, these characteristic indices are used for comprehensive analysis, the comprehensive index of each area is calculated, and it is determined whether it is a partial discharge area. Subsequently, the real-time discharge frequency and repetition rate are compared with the historical normal data to determine whether the partial discharge is abnormal. If an abnormality is found, the system will start the early warning adjustment mechanism. The mechanism includes adjusting parameters such as voltage and temperature to avoid the impact of discharge on the equipment. Through this method, partial discharge can be accurately detected under high noise interference, and equipment failure caused by discharge can be effectively prevented. The present invention not only improves the monitoring accuracy, but also optimizes the emergency response mechanism to ensure the safe operation of power equipment.

[0187] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0188] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0189] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0190] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0191] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of application of the present invention should still fall within the scope of the patent coverage of this invention.

Claims

1. A method for detecting partial discharge of cables under high noise, characterized in that: The following steps are involved: S1: During the monitoring period, the cable is divided into several identical areas, and the current pulse signal in each area is obtained respectively; Wherein, the current pulse signal is collected by a current sensor, including the current pulse frequency and the current pulse repetition rate; S2: preprocessing the current pulse signal in each region, extracting the current pulse frequency fluctuation characteristics and the current pulse repetition rate characteristics, and generating the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index; S3: Comprehensively analyzing the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each area, calculating the comprehensive index of each area, and determining whether the corresponding area is a partial discharge area; S4: comparing and analyzing the real-time discharge frequency and discharge repetition rate of the determined partial discharge area with the historical normal discharge frequency and discharge repetition rate to determine whether the partial discharge is abnormal; S5: According to the judgment result, when partial discharge abnormality is detected, the system starts the early warning adjustment mechanism.

2. A cable partial discharge detection method under high noise according to claim 1, characterized in that: The process of obtaining the current pulse frequency fluctuation characteristic index specifically includes: Obtaining current pulse frequency data in each area; Preprocessing the current pulse frequency data; Identify all local maxima and local minima in the current pulse frequency data; Use spline interpolation to connect local maximum and minimum values ​​to generate upper and lower envelopes; Calculate the mean of the upper and lower envelopes; The first intrinsic mode function is extracted from the current pulse frequency data using the empirical mode decomposition method; Repeat the steps for the remaining signals to continue extracting all the intrinsic mode functions; Perform Hilbert transform on each eigenmode function to obtain the analytical signal; Calculate the instantaneous amplitude and instantaneous phase of the analytical signal; The instantaneous frequency is calculated by taking the derivative of the instantaneous phase with respect to time, and the standard deviation σ of the instantaneous frequency is calculated. f and the average value μ f , the current pulse frequency fluctuation characteristic index is calculated based on the standard deviation and average value of the instantaneous frequency.

3. The cable partial discharge detection method under high noise according to claim 1, characterized in that: The process of obtaining the current pulse repetition rate characteristic index specifically includes: Acquire the current pulse repetition rate data in each region to form a data matrix X; The current pulse repetition rate data is standardized to obtain the standardized data matrix Z. According to the standardized data matrix Z, calculate the covariance matrix C; Perform eigenvalue decomposition on the covariance matrix C to obtain eigenvalues ​​and eigenvectors; Select the first k largest eigenvalues ​​and their corresponding eigenvectors to form the principal component matrix V k ; Where k represents the number of eigenvalues; Project the standardized data into the principal component space to obtain the reduced-dimensional data Z k ; Reconstruct the original data according to the principal component projection results and obtain the reconstructed data matrix Z s ; Calculate the reconstruction error matrix E; Calculate the reconstruction error norm RE for each data point i ; Among them, the calculation expression of the reconstruction error norm is: RE i =||E i ||; In the formula, i represents the number of data points, RE i represents the reconstruction error norm of the i-th data point, E i represents the reconstruction error matrix of the i-th data point; Calculate the mean μ of the reconstruction error norm F and standard deviation σ F , and the mean and standard deviation of the reconstruction error norm are normalized to calculate the current pulse repetition rate characteristic index.

4. The cable partial discharge detection method under high noise according to claim 1, characterized in that: The current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each region are comprehensively analyzed to calculate the comprehensive index of each region, specifically including: The current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each region are obtained, and the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index in each region are normalized to calculate the comprehensive index of each region.

5. The cable partial discharge detection method under high noise according to claim 1, characterized in that: The determining whether the corresponding area is a partial discharge area specifically includes: It is determined whether the comprehensive index is less than a preset comprehensive index threshold. If so, it is recorded as a partial discharge area; if not, it is recorded as a non-partial discharge area.

6. The cable partial discharge detection method under high noise according to claim 1, characterized in that: The real-time discharge frequency and discharge repetition rate of the determined partial discharge area are compared and analyzed with the historical normal discharge frequency and discharge repetition rate to determine whether the partial discharge is abnormal, specifically including: Obtain the historical normal discharge frequency and discharge repetition rate of multiple acquisition points according to the time series, and record them as historical normal data; Standardize historical data; The processed historical data is used as a training set to train the isolation forest model; The average isolation degree of the corresponding historical data points is calculated through the isolation depth of multiple trees; Isolation Forest partitions historical data using multiple trees and ultimately generates an isolation score for each historical data point; The calculation expression of the isolation score is: Where h(x) represents the isolation score of historical data point x, H(x) represents the average isolation depth in the isolation tree of historical data point x, m represents the number of historical data in the data set, and c(m) represents the maximum isolation depth in the isolation tree; The normal discharge frequency and discharge repetition rate of each area are collected in real time according to the time series and recorded as regional real-time data; Standardize real-time data; The processed real-time data of each region is input into the trained isolation forest model for detection; The isolation score of each regional real-time data point is calculated using the isolation forest model; For each region, the average isolation score of all real-time data points is calculated and recorded as the real-time region average isolation score; For each region, the average isolation score of all historical data points is calculated and recorded as the historical region average isolation score; The difference between the real-time regional average isolation score and the historical regional average isolation score is calculated, and the absolute value is removed and recorded as the regional deviation value; comparing the region deviation value to a region deviation value threshold; If the regional deviation value is greater than or equal to the regional deviation value threshold, it is recorded as an abnormal area; If the area deviation value is less than the area deviation value threshold, it is recorded as a normal area.

7. The cable partial discharge detection method under high noise according to claim 1, characterized in that: According to the judgment result, when the partial discharge abnormality is detected, the system starts the early warning adjustment mechanism, which specifically includes: Based on the abnormal area, the system will automatically take alarm response; After the alarm, the system automatically takes control measures, including: Calculate the standard deviation and average value of the voltage in the corresponding area in the period before the warning, calculate the ratio of the standard deviation to the average value of the voltage, and obtain the voltage jump ratio; comparing the voltage jump ratio with a preset threshold; If the voltage jump ratio is greater than or equal to the preset threshold, the real-time voltage is adjusted according to the historical normal voltage range; If the voltage jump ratio is less than the preset threshold, the temperature of the insulation layer is detected to determine whether the temperature of the insulation layer is greater than or equal to the preset temperature threshold. If so, the temperature is lowered through the temperature control system; After adjusting the voltage and temperature of the cable during operation, re-judge the discharge situation in the corresponding area. If abnormal discharge still exists in the area, power-off isolation measures are taken.

8. A cable partial discharge detection system under high noise, characterized in that: A cable partial discharge detection method under high noise conditions as described in any one of claims 1 to 7, comprising: A data acquisition module, wherein the data acquisition module is used to divide the cable into a number of identical areas during a monitoring period and obtain a current pulse signal in each area; A data preprocessing module, which preprocesses the current pulse signal in each area, extracts the current pulse frequency fluctuation characteristics and the current pulse repetition rate characteristics, and generates a current pulse frequency fluctuation characteristic index and a current pulse repetition rate characteristic index; A partial discharge area determination module, wherein the partial discharge area determination module calculates a comprehensive index of each area by comprehensively analyzing a current pulse frequency fluctuation characteristic index and a current pulse repetition rate characteristic index in each area, and determines whether the corresponding area is a partial discharge area; A partial discharge abnormality analysis module, which is used to compare and analyze the real-time discharge frequency and discharge repetition rate of the determined partial discharge area with the historical normal discharge frequency and discharge repetition rate to determine whether the partial discharge is abnormal; The early warning adjustment module starts the early warning adjustment mechanism according to the judgment result when the partial discharge abnormality is detected.

9. A cable partial discharge detection system under high noise according to claim 8, characterized in that: The data preprocessing module at least includes: A current pulse frequency fluctuation characteristic index calculation unit, wherein the current pulse frequency fluctuation characteristic index calculation unit calculates the current pulse frequency fluctuation characteristic index by processing the current pulse frequency; A current pulse repetition rate characteristic index calculation unit is used to calculate the current pulse repetition rate characteristic index by processing the current pulse repetition.

10. The cable partial discharge detection system under high noise according to claim 8, characterized in that: The partial discharge area determination module at least includes: A comprehensive index calculation unit is provided, wherein the comprehensive index is obtained by comprehensively calculating the current pulse frequency fluctuation characteristic index and the current pulse repetition rate characteristic index.

Images