Partial discharge real-time monitoring system based on pulse current

Abstract

The invention relates to the technical field of partial discharge monitoring, and discloses a partial discharge real-time monitoring system based on pulse current, which comprises a signal acquisition module, a preprocessing module, a preliminary diagnosis module, a fault isolation module and a deep diagnosis module. Interference is accurately removed, and data reliability is guaranteed; faults are quickly positioned through targeted detection of open circuit of the sensor and abnormity of the sampling card. The fault isolation module can automatically switch a standby sensor when a fault occurs, so that the fault influence is reduced; the deep diagnosis module constructs a fault feature vector by using multi-dimensional features, and compares the fault feature vector with a sample set to realize fine fault classification; the fault early warning module can send sensor and partial discharge fault early warning signals to operation and maintenance personnel in real time; the monitoring accuracy and reliability are improved, a powerful guarantee is provided for safe and stable operation of equipment, and the accident occurrence risk is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of partial discharge monitoring, and particularly to a real-time partial discharge monitoring system based on pulse current. Background Art

[0002] In modern power systems, the safe and stable operation of power equipment is directly related to the reliability of the power grid and the order of social production and life. As an early typical sign of insulation deterioration of power equipment, if partial discharge is not detected and processed in time, it may lead to equipment failures or even large-scale power outages. Therefore, real-time monitoring and accurate diagnosis of partial discharge have become key technical requirements in the field of condition-based maintenance of power equipment.

[0003] At present, the partial discharge monitoring technology based on the pulse current method has been widely applied. It captures the pulse current signals generated by partial discharge of equipment through high-frequency current sensors to realize the perception of discharge phenomena. However, there are still many limitations in the existing technology: First, in the signal processing link, factors such as environmental noise and electromagnetic interference easily cause a large number of invalid components to be mixed in the original pulse current signals. Traditional threshold determination or simple filtering methods are difficult to effectively distinguish real discharge pulses from interference signals, reducing the monitoring accuracy; Second, at the level of fault diagnosis, most existing systems only focus on the identification of partial discharge itself and lack the fault diagnosis ability for the hardware of the monitoring system. Once abnormal situations such as open circuits or crashes occur in the hardware, it is easy to cause distortion or interruption of monitoring data, thereby leading to misjudgment or missed judgment; Third, in terms of fault type identification, most systems only rely on a single feature for diagnosis, which is difficult to comprehensively reflect the complex characteristics of partial discharge and cannot accurately distinguish different types of discharge faults, limiting the depth and reliability of equipment condition assessment.

[0004] Therefore, there is an urgent need for a real-time partial discharge monitoring system that can achieve efficient preprocessing of signals, collaborative diagnosis of hardware faults and discharge faults, and multi-dimensional accurate identification of fault types to overcome the deficiencies of the existing technology and improve the intelligent and reliable level of operation and maintenance of power equipment. Summary of the Invention

[0005] The purpose of the present invention is to provide a real-time partial discharge monitoring system based on pulse current, which solves the technical problems raised in the background art.

[0006] The purpose of the present invention can be achieved by the following technical solutions: A real-time partial discharge monitoring system based on pulse current, comprising: A signal acquisition module, configured to sense the pulse current signals generated by partial discharge in real time through a high-frequency current sensor; A preprocessing module, configured to preprocess the pulse current signals to determine the effective pulses at relevant acquisition time nodes; The preliminary diagnosis module is used to perform fault diagnosis and analysis based on the pulsed current signal, so as to determine whether there is an open circuit fault in the high-frequency current sensor and an abnormality in the sampling card; The fault isolation module is used to receive the fault identification sent by the preliminary diagnosis module, then trigger the corresponding preset emergency handling mechanism according to the fault identification, and determine the sensor fault warning signal; When it is determined by the preliminary diagnosis module that there is no open circuit fault in the high-frequency current sensor and the sampling card is normal, the deep diagnosis module identifies the fault type of partial discharge, so as to determine the partial discharge fault warning signal; The fault warning module is used to send the sensor fault warning signal and the partial discharge fault warning signal to the relevant operation and maintenance personnel; As a further solution of the present invention: the preprocessing method is as follows: Step S1, data marking: Mark the pulsed current signal generated by the partial discharge sensed in real time by the high-frequency current sensor as L t , where t is the serial number of the acquisition time node of the pulsed current signal; Within the pre-specified time window C1, obtain the pulsed current signals at multiple acquisition time nodes therein, and record them as L k , k = 1, 2,... v, where v is the number of multiple acquisition time nodes within the pre-specified time window C1; Step S2, baseline calibration: By: , calculate the pulsed current signal L0 after baseline calibration t ; First, obtain the average value of multiple pulsed current signals within the pre-specified time window, and then subtract the average value from the pulsed current signal L t at the corresponding acquisition time node, then the pulsed current signal after baseline calibration can be obtained; Step S3, pulse extraction: Extract the pulsed current signals L0 k and L0 k+1 at two adjacent acquisition time nodes, and the time interval T between them; Then by: , calculate the signal amplitude change rate R between them t ; Extract the preset amplitude change threshold R y and the lowest signal threshold L min ; When |R t | > R y , and L0 t > L min , it is determined that the pulsed current signal at the t-th acquisition time node is a valid pulse; Otherwise, it is determined that the pulsed current signal at the t-th acquisition time node does not belong to a valid pulse.

[0007] As a further solution of the present invention, the fault diagnosis and analysis method is as follows: Step G1, sensor open circuit detection: When it is determined that the pulsed current signal at the t-th acquisition time node does not belong to a valid pulse, the pulsed current signals at n consecutive acquisition time nodes after the t-th acquisition time node are obtained, and the baseline voltage is extracted at the same time: If all the pulsed current signals at the n acquisition time nodes do not belong to valid pulses and the baseline voltage value is 0, the high-frequency current sensor is determined to have an open circuit fault, and a fault identifier "A1" is added to it accordingly; Otherwise, the high-frequency current sensor is not determined to have an open circuit fault, and a fault identifier "A0" is added to it accordingly; Among them, the baseline voltage is used to describe the reference level or stable state voltage value of the pulsed current signal when there is no pulse; Step G2, sampling card anomaly detection: Within a pre-specified time window C2, the pulsed current signals at multiple acquisition time nodes are obtained and denoted as L j , j = 1, 2,... m, where m is the number of multiple acquisition time nodes within the pre-specified time window C2; Through the formula , the variance BL corresponding to L j is calculated; Among them, L j is the pre-processed pulsed current signal, and the variance BL is used to measure the fluctuation degree of the pulsed current signal, and PL is the average value of all L j ; The pre-set crash threshold BL is extracted y ; When BL < BL y , it is determined that the sampling card is abnormal, and a fault identifier "B1" is added to it accordingly; Otherwise, the high-frequency current sensor is not determined to be abnormal for the sampling card, and a fault identifier "B0" is added to it accordingly; As a further solution of the present invention, the emergency handling mechanism is as follows: When the fault identifier is "A1" or "B1", the high-frequency current sensor is automatically switched to the standby one, and a sensor fault warning signal is generated; As a further solution of the present invention, the fault type identification method is as follows: Step H1, pulse repetition rate feature: Within a pre-specified time window C3, the number g of valid pulses is counted; Pass: Calculate the pulse repetition rate CF within the time window C3; Wherein, T3 is the duration of the time window C3; Step H2, Pulse amplitude mean feature: Extract the pulse current signal L corresponding to all valid pulses within the time window C3 i , where i is the acquisition time node number corresponding to the valid pulse within the time window C3; Pass: Calculate the pulse amplitude mean AL within the time window C3; Step H3, Pulse rise time mean feature: For each valid pulse, the time experienced from rising from 10% of the pulse starting amplitude to 90% of the amplitude is defined as the pulse rise time; Within the time window C3, calculate the average value of the rise times of all valid pulses, and denote it as the pulse rise time mean ST; Step H4, Fault type determination: Take the pulse repetition rate CF, the pulse amplitude mean AL, and the pulse rise time mean ST as the fault feature vector W to be judged = {CF, AL, ST}; At the same time, extract the fault feature vector sample set WY corresponding to different typical fault types obtained through experiments in advance u = {CF u , AL u , ST u}, where u = 1, 2,..., q, and q represents the number of fault feature vector samples corresponding to different typical fault types; Among them, CF u is the pulse repetition rate of the u-th sample, AL u is the pulse amplitude mean of the u-th sample, and ST u is the pulse rise time mean of the u-th sample; Then pass:

[0008] Calculate the distance D between the fault feature vector W to be judged and the fault feature vector samples WY corresponding to different typical fault types u ; u ; Compare the magnitudes of all distances D u and obtain the minimum distance D u，min ; When D u，min ≤ Dy, it is determined that the current partial discharge fault type is the same as the fault type corresponding to the fault feature vector sample WY u , and at the same time, generate a partial discharge fault warning signal; Among them, Dy is a preset distance threshold.

[0009] Advantages of the present invention: Precise signal acquisition and preprocessing: By using a high-frequency current sensor to capture partial discharge pulse current signals in real time, and adopting preprocessing methods such as data marking, baseline calibration, and pulse extraction, signal interference and baseline drift are effectively removed. Effective pulses are accurately determined based on the signal amplitude change rate and the lowest signal threshold, providing a reliable data basis for subsequent fault diagnosis, and significantly improving the accuracy and effectiveness of signal acquisition.

[0010] Efficient fault diagnosis and isolation: Fault diagnosis and analysis cover sensor open-circuit detection and sampling card anomaly detection. Scientific judgment rules are set respectively based on the continuity of effective pulses and signal variance, enabling rapid identification of high-frequency current sensor open-circuit faults and sampling card anomalies. After receiving the fault identification, the fault isolation module quickly triggers an emergency handling mechanism, automatically switches to a standby sensor, reduces monitoring interruptions caused by faults, ensures the stable operation of the system, and reduces the scope of fault impact.

[0011] Deep fault type identification: The deep diagnosis module constructs a fault feature vector using multi-dimensional features such as pulse repetition rate, average pulse amplitude, and average pulse rise time, compares it with a typical fault type sample set obtained through prior experiments, and determines the fault type by calculating the distance, realizing refined classification of partial discharge faults, providing more detailed fault information for maintenance personnel, helping to formulate targeted maintenance strategies, and improving fault handling efficiency.

[0012] Real-time and comprehensive fault warning: The fault warning module promptly sends sensor fault warning signals and partial discharge fault warning signals to maintenance personnel, enabling maintenance personnel to grasp the system operation status in real time, quickly respond when a fault occurs or is about to occur, avoid the expansion of the fault causing serious accidents, ensure the safe and stable operation of the equipment, and improve the reliability and safety of application scenarios such as power systems. Description of the drawings

[0013] The present invention will be further described below with reference to the drawings.

[0014] Figure 1 is a system block diagram of a partial discharge real-time monitoring system based on pulse current of the present invention.

[0015] Figure 2 is a schematic flow chart of a preprocessing module in a partial discharge real-time monitoring system based on pulse current of the present invention.

[0016] Figure 3 is a schematic flow chart of a deep diagnosis module in a partial discharge real-time monitoring system based on pulse current of the present invention. Detailed implementation manners

[0017] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0018] As the first embodiment of the present invention: Please refer to Figure 1 、 Figure 2 and Figure 3 As shown, the present invention is a real-time partial discharge monitoring system based on pulse current, including: A signal acquisition module, configured to sense the pulse current signal generated by partial discharge through a high-frequency current sensor, and convert the analog signal corresponding to the pulse current signal into a digital signal through an analog-to-digital converter; At the same time, the pulse current signal converted into a digital signal is marked as L t , where t is the acquisition time node serial number of the pulse current signal; A preprocessing module, configured to perform preprocessing on the converted digital signal; In this embodiment, the purpose of setting the preprocessing module is: since the actual acquisition environment often has various interference factors, such as electromagnetic interference, environmental noise, etc., these interferences will cause the collected original pulse current signal to be doped with a large amount of useless information, thus affecting the subsequent accurate judgment of the partial discharge signal. Therefore, it is necessary to preprocess the signal; The preprocessing method is as follows: Step S1, data marking: Within a pre-specified time window C1, obtain the pulse current signals at multiple acquisition time nodes therein, and record them as L k , where k = 1, 2,..., v, and v is the number of multiple acquisition time nodes within the pre-specified time window C1; Step S2, baseline calibration: By: , calculate the pulse current signal L0 after baseline calibration t ; In this embodiment, the purpose of baseline calibration is to eliminate environmental noise interference. The noise interference in the environment is usually random and continuously present, which will cause the collected signal to deviate from the true partial discharge signal as a whole. Through baseline calibration, the DC offset and low-frequency noise in the signal can be effectively removed, making the subsequent analyzed signal purer and more capable of reflecting the true situation of partial discharge; Step S3, pulse extraction: Extract the pulsed current signals L0 at two adjacent acquisition time nodes k and L0 k+1 , and the time interval T therebetween; Subsequently, through: , calculate the signal amplitude change rate R therebetween t ; Extract the preset amplitude change threshold R y and the lowest signal threshold L min ; When |R t | > R y , and L0 t > L min , it is determined that the pulsed current signal at the t-th acquisition time node is a valid pulse; Otherwise, it is determined that the pulsed current signal at the t-th acquisition time node does not belong to a valid pulse; Preliminary diagnosis module, used for fault diagnosis and analysis based on the pulsed current signal: The specific method is as follows: Step G1, sensor open circuit detection: When it is determined that the pulsed current signal at the t-th acquisition time node does not belong to a valid pulse, obtain the pulsed current signals at n consecutive acquisition time nodes after the t-th acquisition time node, and simultaneously extract the baseline voltage: If all the pulsed current signals at the n acquisition time nodes do not belong to valid pulses, and the baseline voltage value is 0; then the high-frequency current sensor is determined to have an open circuit fault, and a fault identifier "A1" is added to it; Otherwise, the high-frequency current sensor is not determined to have an open circuit fault, and a fault identifier "A0" is added to it; Among them, the baseline voltage is used to describe the reference level or stable state voltage value of the pulsed current signal when there is no pulse; Step G2, sampling card abnormality detection: Within the pre-specified time window C2, obtain the pulsed current signals at multiple acquisition time nodes therein, and denote them as L j , j = 1, 2,... m, where m is the number of multiple acquisition time nodes within the pre-specified time window C2; Through the formula , calculate the variance BL corresponding to L j ; Among them, L j is the preprocessed pulsed current signal, the variance BL is used to measure the signal fluctuation degree of the pulsed current signal, and PL is the average value of all L j ; Extract the preset crash threshold BL y ; When BL < BL y , it is determined that the sampling card is abnormal, and then a fault identifier "B1" is added to it; On the contrary, the high-frequency current sensor is not determined to be abnormal in the sampling card, and then a fault identifier "B0" is added to it; In this embodiment, under normal circumstances, the pulsed current signal will have a certain fluctuation, and the variance will not be too small. If the variance is too small, it means that there may be a problem with the sampling card, resulting in the pulsed current signal not fluctuating normally; For example, the sampling card crashing may cause the collected pulsed current signal to be fixed near a certain value with little change, resulting in a very small variance; The fault isolation module is used to receive the fault identifier sent by the preliminary diagnosis module, and then trigger the corresponding preset emergency handling mechanism according to the fault identifier; The emergency handling mechanism is as follows: When the fault identifier is "A1" or "B1", the standby high-frequency current sensor is automatically switched, and a sensor fault warning signal is sent through the fault warning module; During the monitoring process of power equipment, sensor faults may occur at any time. If not handled in time, it will lead to the loss of monitoring data, affecting the accurate judgment of the equipment operation status. Therefore, in this embodiment, when a fault such as a sensor open circuit or a sampling card crash is detected, the system immediately enables the standby sensor; Among them, the sensor fault warning signal is used to notify the relevant operation and maintenance personnel so that the operation and maintenance personnel can repair or replace the faulty sensor in time according to the fault identifier corresponding to the sensor fault warning signal; The partial discharge real-time monitoring system based on pulsed current constructed in Embodiment 1 accurately senses and converts the pulsed current signal through the signal acquisition module, laying a data foundation for subsequent processing. The preprocessing module specifically adopts data marking, baseline calibration, and pulse extraction methods to effectively eliminate useless information generated by factors such as electromagnetic interference and environmental noise, significantly improving the purity and reliability of the signal. The preliminary diagnosis module can detect faults such as high-frequency current sensor open circuit and sampling card abnormality in a timely manner, and the fault isolation module can quickly start the standby sensor and issue a warning according to the fault identifier, greatly reducing the problem of missing monitoring data caused by sensor faults, ensuring the continuity and stability of the power equipment monitoring process, providing a strong guarantee for the safe operation of power equipment, and effectively reducing the potential risks caused by monitoring failure.

[0019] As Embodiment 2 of the present invention: Please refer to Figure 1 , Figure 2 and Figure 3As shown, when the present application is specifically implemented, compared with the first embodiment, the difference between the technical solution of this embodiment and that of the first embodiment is only that in this embodiment, it further includes: A deep diagnosis module, which, after determining that the high-frequency current sensor has no open circuit fault and the sampling card is normal through the preliminary diagnosis module, identifies the fault type of partial discharge; The specific method is as follows: Step H1, Pulse Repetition Rate Feature: Within a pre-specified time window C3, count the number g of valid pulses; Through: , calculate the pulse repetition rate CF within the time window C3; In the formula, T3 is the duration of the time window C3; Step H2, Pulse Amplitude Mean Feature: Extract the pulse current signal L corresponding to all valid pulses within the time window C3 i , where i is the acquisition time node serial number corresponding to the valid pulse within the time window C3; Through: , calculate the pulse amplitude mean AL within the time window C3; Step H3, Pulse Rise Time Mean Feature: For each valid pulse, the time experienced from 10% of the pulse starting amplitude to 90% of the amplitude is defined as the pulse rise time; Within the time window C3, calculate the average value of the rise times of all valid pulses and denote it as the pulse rise time mean ST; In this embodiment, the pulse rise time is related to the physical mechanism during the discharge process. For example, the pulse rise times of internal discharge and surface discharge may be different; Step H4, Fault Type Judgment: Take the pulse repetition rate CF, the pulse amplitude mean AL, and the pulse rise time mean ST as the fault feature vector W to be judged = {CF, AL, ST}; At the same time, extract the fault feature vector sample set WY corresponding to different typical fault types obtained through experiments in advance u = {CF u , AL u , ST u}, where u = 1, 2,..., q, and q represents the number of fault feature vector samples corresponding to different typical fault types; In this embodiment, typical fault types include insulation air gap discharge, insulation surface discharge, insulation internal discharge, etc.; Then through:

[0020] Calculate the distance D between the fault feature vector W to be judged and the fault feature vector samples WY corresponding to different typical fault types u ; u ; Compare all the distances D u in terms of their magnitudes, and obtain the distance D with the minimum value u，min ; When D u，min ≤ Dy, it is determined that the current partial discharge fault type is the same as the fault type corresponding to the fault feature vector sample WY u , and at the same time, a partial discharge fault warning signal is generated; wherein, Dy is a preset distance threshold, and in this embodiment, the value of Dy approaches 0; The fault warning module is further configured to identify the faulty device on the display interface of the monitoring system with a preset color according to the partial discharge fault warning signal, display the current partial discharge fault type, and record detailed information such as the fault occurrence time and device number in the system log for the operation and maintenance personnel to consult.

[0021] Based on Embodiment 1, Embodiment 2 introduces a deep diagnosis module, which uses features such as pulse repetition rate, average pulse amplitude, and average pulse rise time to accurately identify the partial discharge fault type. By comparing and analyzing the fault feature vector to be judged with the pre-obtained fault feature vector sample set corresponding to different typical fault types, this method can accurately determine various fault types such as insulation air gap discharge, insulation surface discharge, and insulation internal discharge, and generate a detailed partial discharge fault warning signal. It not only identifies the faulty device on the monitoring system display interface with a specific color, but also records information such as the fault occurrence time and device number in detail. This enables the operation and maintenance personnel to quickly and accurately grasp the fault situation, greatly improves the accuracy and efficiency of fault diagnosis, provides a precise basis for carrying out equipment maintenance work targeted, helps prevent the deterioration of equipment faults in advance, reduces the risk of equipment damage, and improves the safety and reliability of the power system operation.

[0022] As Embodiment 3 of the present invention: Please refer to Figure 1 , Figure 2 and Figure 3 shown. In the specific implementation of this application, compared with Embodiment 1 and Embodiment 2, the technical solution of this embodiment lies in combining the solutions of the above Embodiment 1 and Embodiment 2 for implementation.

[0023] Example 3 combines the solutions of Example 1 and Example 2 organically. It not only has the advantages of Example 1 in quickly detecting and isolating sensor faults and ensuring the continuity of monitoring, but also has the characteristics of Example 2 in accurately identifying the types of partial discharge faults and providing detailed fault information. This comprehensive and in-depth monitoring and diagnosis method can comprehensively ensure the operation safety of power equipment. On the one hand, it can timely handle sensor faults and avoid interruption of monitoring data; on the other hand, it can accurately identify the types of partial discharge faults, provide detailed fault warnings for maintenance personnel, enable them to efficiently formulate maintenance strategies, comprehensively improve the operation and maintenance management level of power equipment, effectively reduce the probability of power equipment failures, ensure the stable and efficient operation of the power system, and provide double guarantees for the reliability and stability of power supply.

[0024] It should be stated that: all data collected in this application are collected with the consent and authorization of users, and the uses of the data are legal and compliant, and the use and processing of the data comply with the relevant laws, regulations and standards of the relevant regions.

[0025] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by software simulation of a large amount of collected data to get a formula closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0026] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claimed rights.

Claims

1. A real-time partial discharge monitoring system based on pulsed current, characterized in that, Including: A signal acquisition module, which is used to sense the pulse current signal generated by partial discharge in real time through a high-frequency current sensor; A preprocessing module, which is used to preprocess the pulse current signal to determine the effective pulses at relevant acquisition time nodes; A preliminary diagnosis module, which is used to conduct fault diagnosis and analysis based on the pulse current signal to determine whether there is an open circuit fault in the high-frequency current sensor and an abnormality in the sampling card; A fault isolation module, which is used to receive the fault identifier sent by the preliminary diagnosis module, then trigger the corresponding preset emergency processing mechanism according to the fault identifier, and determine the sensor fault warning signal; A deep diagnosis module, when it is determined by the preliminary diagnosis module that there is no open circuit fault in the high-frequency current sensor and no abnormality in the sampling card, it identifies the fault type of the partial discharge to determine the partial discharge fault warning signal.

2. The on-line partial discharge monitoring system based on pulse current according to claim 1, wherein The preprocessing method is as follows: Step S1: Mark the pulse current signal generated by real-time sensing of partial discharge through the high-frequency current sensor as L t , where t is the serial number of the acquisition time node of the pulse current signal; Within a pre-specified time window C1, obtain the pulsed current signals at multiple acquisition time nodes therein and denote them as L k , where k = 1, 2, …… v, and v is the number of multiple acquisition time nodes within the pre-specified time window C1; Step S2: First, calculate the average value of multiple pulsed current signals within a pre-specified time window, and then subtract the average value from the pulsed current signal L t at the corresponding acquisition time node, thereby obtaining the pulsed current signal after baseline calibration; Step S3: Extract the pulsed current signals L0 at two adjacent acquisition time nodes k and L0 k+1 , and the time interval T therebetween; Subsequently, pass through: , and calculate the signal amplitude change rate R between them t ; Extract the preset amplitude change threshold R y and the lowest signal threshold L min ; When |R t | > R y , and L0 t > L min , then it is determined that the pulsed current signal at the t-th acquisition time node is a valid pulse; Otherwise, it is determined that the pulse current signal at the t-th acquisition time node does not belong to an effective pulse.

3. The on-line partial discharge monitoring system based on pulse current according to claim 2, wherein, The sensor open circuit detection method in the fault diagnosis and analysis is as follows: When it is determined that the pulse current signal at the t-th acquisition time node does not belong to an effective pulse, the pulse current signals at n consecutive acquisition time nodes after the t-th acquisition time node are obtained, and the baseline voltage is extracted at the same time: If all the pulse current signals at the n acquisition time nodes do not belong to effective pulses and the baseline voltage value is 0; then the high-frequency current sensor is determined to have an open circuit fault, and a fault identifier "A1" is added to it; Otherwise, the high-frequency current sensor is not determined to have an open circuit fault, and a fault identifier "A0" is added to it; Among them, the baseline voltage is used to describe the reference level of the pulse current signal when there is no pulse.

4. A real-time partial discharge monitoring system based on pulsed current according to claim 3, characterized in that, The sampling card abnormality detection method in the fault diagnosis and analysis is as follows: Within a pre-specified time window C2, obtain the pulsed current signals at multiple acquisition time nodes therein and denote them as L j , where j = 1, 2, …… m, and m is the number of multiple acquisition time nodes within the pre-specified time window C2; Calculate L through the formula and obtain the variance BL corresponding to L j ​ Among them, L j is the preprocessed pulse current signal, the variance BL is used to measure the fluctuation degree of the pulse current signal, and PL is the average value of all L j ; Extract a preset crash threshold BL y ; When BL < BL y , it is determined that the sampling card is abnormal, and then a fault identifier "B1" is added to it; Otherwise, the high-frequency current sensor is not determined to have an abnormality in the sampling card, and a fault identifier "B0" is added to it.

5. The real-time partial discharge monitoring system based on pulse current according to claim 4, characterized in that, The emergency processing mechanism is as follows: When the fault identifier is "A1" or "B1", the high-frequency current sensor is automatically switched to the standby one, and a sensor fault warning signal is generated.

6. The on-line partial discharge monitoring system based on pulse current according to claim 1, characterized in that, The fault type identification method is as follows: Extract the fault feature vector W = {CF, AL, ST} to be judged for partial discharge; Among them, CF represents the pulse repetition rate, AL represents the average pulse amplitude, and ST represents the average pulse rise time; Simultaneously extract the fault feature vector sample set WY corresponding to different typical fault types obtained through experiments in advance u = {CF u , AL u , ST u}, where u = 1, 2, …… q, and q represents the number of fault feature vector samples corresponding to different typical fault types; Among them, CF u is the pulse repetition rate of the u-th sample, AL u is the mean pulse amplitude of the u-th sample, and ST u is the mean pulse rise time of the u-th sample; Followed by: ; Calculate the distance D between the fault feature vector W to be judged and the fault feature vector samples WY corresponding to different typical fault types u ; u ; Compare all distances D u in terms of magnitude, and obtain the distance D with the minimum value u，min ; When D u，min ≤ Dy, it is determined that the current partial discharge fault type is the same as the fault type corresponding to the fault feature vector sample WY u Meanwhile, a partial discharge fault warning signal is generated; where Dy is a preset distance threshold.

7. A real-time partial discharge monitoring system based on pulsed current according to claim 6, characterized in that The calculation method of the pulse repetition rate is: Within the pre-specified time window C3, count the number g of effective pulses; Adopted by: , the pulse repetition rate CF within the time window C3 is calculated; In the formula, T3 is the duration of the time window C3.

8. A real-time partial discharge monitoring system based on pulse current according to claim 6, characterized in that, The calculation method of the average pulse amplitude is: Extract the pulse current signal L corresponding to all valid pulses within the time window C3 i , where i is the serial number of the acquisition time node corresponding to the valid pulse within the time window C3; Adopted by: , the average pulse amplitude AL within the time window C3 is calculated.

9. A real-time partial discharge monitoring system based on pulse current according to claim 6, characterized in that, The calculation method of the average pulse rise time is: For each effective pulse, the time experienced from 10% of the pulse starting amplitude to 90% of the amplitude is defined as the pulse rise time; Within the time window C3, calculate the average value of the rise times of all effective pulses, which is the average pulse rise time ST.

10. A real-time partial discharge monitoring system based on pulsed current according to claim 1, characterized in that, It also includes: A fault warning module, which is used to send the sensor fault warning signal and the partial discharge fault warning signal to relevant operation and maintenance personnel.

Images