Capacitive coupling and ultrahigh frequency detection combined partial discharge monitoring system and method

Abstract

The invention belongs to the technical field of partial discharge monitoring, and particularly relates to a capacitive coupling and ultrahigh frequency detection combined partial discharge monitoring system and method.The capacitive coupling and ultrahigh frequency detection combined partial discharge monitoring system comprises a pulse current data module and an ultrahigh frequency data acquisition module, and a capacitive coupling sensor is used for acquiring monitoring data; the data processing and analyzing module is used for processing and analyzing the partial discharge signal; the alarm module outputs an alarm signal to the cloud platform according to the analysis result: when the analysis result shows that partial discharge is suspected to exist, the alarm module outputs the alarm signal; the ultrahigh frequency sensor is used for capturing electromagnetic wave signals generated by partial discharge; when the analysis result is suspected partial discharge, the ultrahigh frequency sensor calculates the specific position where partial discharge occurs based on the time difference generated by the electromagnetic wave signal; a pulse current sensor and an ultrahigh frequency sensor are combined, when a pulse current signal shows that partial discharge is suspected to exist, the ultrahigh frequency sensor verifies whether the partial discharge really exists or not, and positioning is achieved; otherwise, the signals belong to interference signals.

Description

Technical Field

[0001] The invention belongs to the technical field of partial discharge monitoring, and in particular is a partial discharge monitoring system and method combining capacitive coupling with ultra-high frequency detection. Background Art

[0002] With the continuous progress and evolution of the power sector, the power grid system has become larger and more complex. The application of intelligent devices has brought significant advantages to the operation of the power grid. Among them, the ring main unit, as an advanced power equipment, occupies a leading position in the field of high-voltage switch technology. The ring main unit is an important device mainly used for power distribution and control in the power system. The main uses of the ring main unit include improving power supply reliability, power distribution control, protection and isolation, etc. The ring main unit is a high-voltage switchgear, usually installed in a metal or non-metallic insulating cabinet, and the core part consists of a load switch and a fuse. It has a simple structure, small size, low cost, and can improve power supply parameters and performance to ensure power supply safety; however, the ring main unit is prone to various defects during assembly, manufacturing, transportation, and operation aging. These defects will gradually increase over time and may cause insulation breakdown or flashover failures. It is worth noting that most potential faults inside the ring main unit will cause partial discharge in the early stage. By monitoring these partial discharge signals, the insulation degradation or early insulation failure of the equipment can be detected in time, thereby preventing the fault from further deteriorating and ensuring the stable and safe operation of the power grid equipment.

[0003] In power systems, partial discharge detection is of great significance for evaluating the insulation condition of electrical equipment and preventing equipment failures. Using high-voltage sensors in the cabinet as coupling capacitors for pulse current partial discharge detection can capture weak partial discharge signals. However, although this method has high sensitivity and can detect partial discharge in time, it cannot determine the specific location of partial discharge, which limits its application in comprehensive diagnosis and maintenance of power equipment. However, when using only UHF sensors for partial discharge monitoring, it is impossible to correlate the signal amplitude with the severity of partial discharge due to different monitoring mechanisms. In addition, due to the monitoring frequency limitation of the UHF sensor, some partial discharge conditions may be ignored, resulting in the partial discharge conditions corresponding to the electromagnetic wave signals captured by the UHF sensor at a later stage being very severe, which may even cause equipment damage or power outages. To this end, the present invention provides a partial discharge monitoring system and method combining capacitive coupling with ultra-high frequency detection. Summary of the invention

[0004] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0005] The technical solution adopted by the present invention to solve the technical problem is: a partial discharge monitoring system and method combining capacitive coupling and ultra-high frequency detection described in the present invention comprises: A data acquisition module, including a pulse current sensor, for collecting monitoring data; A data processing and analysis module, used for processing and analyzing pulse current signals; The alarm module outputs alarm signals to the cloud platform based on the analysis results: When the analysis result shows that partial discharge is suspected, an alarm signal is output; The data acquisition module also includes: Ultra-high frequency sensor, used to capture electromagnetic wave signals generated by partial discharge; When the analysis result is suspected partial discharge, the UHF sensor calculates the specific location where the partial discharge occurs based on the time difference generated by the electromagnetic wave signal.

[0006] Preferably, the method in which the alarm module outputs an alarm signal according to the analysis result is: The data acquisition module collects and processes the pulse current signal from the pulse current sensor; comparing the processed pulse current signal with a threshold value; When the value of the pulse current signal is greater than the threshold, a judgment that a partial discharge is suspected to exist is output; When the value of the pulse current signal is less than or equal to the threshold value, the judgment that a partial discharge is suspected to exist is not output.

[0007] Preferably, the pulse current signal includes a pulse amplitude U, a pulse count J and a pulse phase M; the threshold is a preset value, according to the formula: ,in, Indicates the value of the pulse current signal; , , are the weights of pulse amplitude U, pulse count J and pulse phase M respectively.

[0008] Preferably, based on the judgment that partial discharge is suspected to exist, the data acquisition module also collects electromagnetic wave signals from the ultra-high frequency sensor, and processes the pulse current signal and the electromagnetic wave signal through the data processing and analysis module, and verifies whether partial discharge exists by extracting the characteristics of the pulse current signal and the electromagnetic wave signal.

[0009] Preferably, the method for verifying whether partial discharge exists is: Signal synchronization, obtaining electromagnetic wave signals and pulse current signals with time stamp synchronization; Signal denoising: denoising electromagnetic wave signals; Feature extraction, extracting key features from electromagnetic wave signals, including frequency distribution and energy spectral density; Signal comparison, compare and analyze the extracted key features with the features in the pulse current signal: If similar features exist, it is determined that partial discharge exists; otherwise, it does not exist.

[0010] Preferably, it also includes a model building module, which is connected to the data processing module and builds a support vector machine model based on the processed pulse current signal and electromagnetic wave signal.

[0011] Preferably, the support vector machine model is used to eliminate interference signals by: According to the formula: ,in, represents the support vector machine model; represents the weight vector, Represents the feature vector after kernel function mapping; represents the bias term; Extract features of the electromagnetic wave signal at the current moment and construct a feature vector X; Input the current feature vector X into the support vector machine model: like If it is a positive number, it is judged that there is partial discharge at the current moment, otherwise it is judged as an interference signal.

[0012] Preferably, the data acquisition module also includes a carbon monoxide sensor for collecting carbon monoxide concentration for verifying partial discharge.

[0013] Preferably, the method for calculating the time difference and determining the specific occurrence location of the partial discharge is: Calculate the time difference between the electromagnetic wave signal and the pulse current signal; According to the formula: ,in, represents the time difference; ti and tj represent the capture time of electromagnetic wave signals at different positions respectively; Based on the time difference, triangulation is used to determine the specific location where the partial discharge occurs.

[0014] A partial discharge monitoring method combining capacitive coupling and ultra-high frequency detection comprises the following steps: S1: The pulse current sensor collects the pulse current signal; S2: Collect electromagnetic wave signals by ultra-high frequency sensor; S3: Processing the pulse current signal and electromagnetic wave signal with the same timestamp; S4: Compare the key features of the pulse current signal and the electromagnetic wave signal to see if they are consistent; S5: Based on the comparison of key features, determine whether partial discharge actually exists. If so, use ultra-high frequency sensors to locate it.

[0015] The beneficial effects of the present invention are as follows: The present invention discloses a partial discharge monitoring system and method combining capacitive coupling with ultra-high frequency detection. The pulse current sensor can capture the sharp current pulses generated during partial discharge. After capturing the pulse current signal, the pulse current signal is processed and compared with the threshold. When the value of the pulse current signal is greater than the threshold, it indicates that partial discharge is suspected to exist. The ultra-high frequency sensor verifies whether the partial discharge actually exists and locates the partial discharge. Otherwise, it is an interference signal. When the value of the pulse current signal is greater than the threshold and the electromagnetic wave signal captured by the ultra-high frequency sensor verifies that the partial discharge actually exists, the alarm unit outputs an alarm signal and notifies the maintenance personnel to perform maintenance and repair on the ring network cabinet where the partial discharge occurs to prevent the partial discharge from worsening, resulting in the insulation material being broken down and causing equipment damage or power outage. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will be further described below in conjunction with the accompanying drawings.

[0017] Figure 1 It is a perspective view of the present invention. DETAILED DESCRIPTION

[0018] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0019] like Figure 1 As shown, a partial discharge monitoring system and method combining capacitive coupling and ultra-high frequency detection described in an embodiment of the present invention include: a data acquisition module, including a pulse current sensor, for collecting monitoring data; a data processing and analysis module, for processing and analyzing pulse current signals; an alarm module, which outputs an alarm signal to a cloud platform according to the analysis result: when the analysis result is suspected partial discharge, an alarm signal is output; the data acquisition module also includes: an ultra-high frequency sensor, for capturing electromagnetic wave signals generated by partial discharge; when the analysis result is suspected partial discharge, the ultra-high frequency sensor calculates the specific location of the partial discharge based on the time difference generated by the electromagnetic wave signal.

[0020] High-frequency pulse currents are generated during partial discharge. By coupling these current signals to the sensor, partial discharge can be detected. When local discharge occurs, the rapid release of charge forms a sharp current pulse. By measuring the size of this pulse, the intensity of the partial discharge can be inferred. The high-voltage sensor in the cabinet acts as a coupling device that can convert these pulse signals into measurable electrical signals, thereby achieving high-sensitivity detection.

[0021] By setting a threshold, when the value of the pulse current signal exceeds the threshold, the analysis result is suspected partial discharge, and the alarm module outputs an alarm signal to the cloud platform and notifies the maintenance personnel.

[0022] Although this method has high sensitivity, it still faces some challenges in practical application. The main problem is that this method cannot directly locate the specific location where partial discharge occurs. Since the signal is captured within the range of sensor coupling, and there may be multiple potential partial discharge points and environmental noise interference in a ring network cabinet, it is difficult to accurately determine the specific discharge location based on the pulse current method alone. This limitation reduces the targetedness of detection work and the efficiency of subsequent maintenance work.

[0023] Based on the above, as an embodiment of the present invention, an ultra-high frequency sensor is set up, and the ultra-high frequency sensor is used to detect the ultra-high frequency electromagnetic waves generated by local discharge, and its output contains information about the discharge position and intensity; the ultra-high frequency sensor mainly monitors electromagnetic wave signals in the frequency range of 500KHz to 1500MHz. This frequency band is less affected by interference, which helps to improve the detection accuracy; after capturing the electromagnetic wave signal, it is processed by denoising and the like, and the existence of the local discharge is verified by extracting features, and then the time difference of the electromagnetic wave signal is captured by multiple ultra-high frequency sensors. The position of the local discharge can be accurately located by using triangulation and multi-sensor fusion technology, wherein the premise of determining the position of the local discharge is to verify the existence of the local discharge by extracting features.

[0024] Preferably, the method in which the alarm module outputs an alarm signal according to the analysis result is: The data acquisition module collects and processes the pulse current signal from the pulse current sensor; comparing the processed pulse current signal with a threshold value; When the value of the pulse current signal is greater than the threshold, a judgment that a partial discharge is suspected to exist is output; When the value of the pulse current signal is less than or equal to the threshold value, the judgment that a partial discharge is suspected to exist is not output.

[0025] As an embodiment of the present invention, the pulse current sensor can capture the sharp current pulses generated by partial discharge, and then after capturing the pulse current signal, it processes the pulse current signal and compares it with the threshold value. The threshold value is a preset value obtained by experiment. When the value of the pulse current signal is greater than the threshold value, it indicates that partial discharge is suspected to exist and the position of the partial discharge cannot be located. The ultra-high frequency sensor is then used to verify whether the partial discharge actually exists and locate it; otherwise it is an interference signal; when the value of the pulse current signal is greater than the threshold value and the electromagnetic wave signal captured by the ultra-high frequency sensor verifies that the partial discharge actually exists, the alarm unit outputs an alarm signal and notifies the maintenance personnel to perform maintenance and repair on the ring network cabinet where partial discharge occurs to prevent the partial discharge from worsening, resulting in the insulation material being broken down and causing equipment damage or power outages.

[0026] Preferably, the pulse current signal includes a pulse amplitude U, a pulse count J and a pulse phase M; the threshold is a preset value, according to the formula: ,in, Indicates the value of the pulse current signal; , , are the weights of pulse amplitude U, pulse count J and pulse phase M respectively.

[0027] As an embodiment of the present invention, the information contained in the pulse current signal includes pulse amplitude, pulse count and pulse phase, etc.; wherein the pulse amplitude indicates the intensity of the discharge event. The higher the amplitude, the more intense the discharge event may be; the pulse count indicates the number of discharges recorded within a specific time; frequent discharges may indicate serious local discharge activity; the pulse phase indicates the phase angle at which the discharge event occurs, which helps to identify the discharge mode and possible discharge source; in addition, the pulse width, frequency and energy are also included, which are used to express the severity of local discharge, distinguish the type of local discharge and evaluate the discharge intensity respectively; During the operation of the ring main unit, in order to determine whether there is partial discharge through the pulse current signal, a threshold is set in advance. Through the setting of the threshold, during the actual operation, when any pulse current signal is captured, the corresponding value is substituted into the formula Z for calculation, and the value of the pulse current signal at the current moment is obtained and compared with the threshold, so as to determine the size relationship between the value of the pulse current signal and the threshold, and then output the judgment result; The expression of the threshold is: ,in, , , These are standard values ​​and are set manually.

[0028] Preferably, based on the judgment that partial discharge is suspected to exist, the data acquisition module also collects electromagnetic wave signals from the ultra-high frequency sensor, and processes the pulse current signal and the electromagnetic wave signal through the data processing and analysis module, and verifies whether partial discharge exists by extracting the characteristics of the pulse current signal and the electromagnetic wave signal.

[0029] As an embodiment of the present invention, since the pulse current sensor can only capture pulse signals with lower frequencies, and there are many interference signals during the operation of the ring network cabinet, the risk of false alarm may occur only by comparing the pulse current signal with the threshold. By adding an ultra-high frequency sensor, the position of the partial discharge can be located, and secondly, by capturing the electromagnetic wave signal, after feature extraction, it is compared with the feature of the pulse current signal to verify whether the partial discharge exists, and further reduce the risk of false alarm caused by interference signals. Preferably, the method for verifying whether partial discharge exists is: Signal synchronization, obtaining electromagnetic wave signals and pulse current signals with time stamp synchronization; Signal denoising: denoising electromagnetic wave signals; Feature extraction, extracting key features from electromagnetic wave signals, including frequency distribution and energy spectral density; Signal comparison, compare and analyze the extracted key features with the features in the pulse current signal: If similar features exist, it is determined that partial discharge exists; otherwise, it does not exist.

[0030] Based on the above, as an embodiment of the present invention, signal synchronization is intended to ensure synchronization of monitoring data timestamps of the pulse current sensor and the ultra-high frequency sensor so as to perform accurate comparative analysis; signal denoising is intended to reduce the influence of environmental noise; it is achieved by digital filters or wavelet denoising and other technologies; feature extraction is intended to identify the type and severity of partial discharge; finally, the extracted key features are compared and analyzed with the key features of the pulse current signal. If similar features exist, it can be determined that there is partial discharge, and then multiple ultra-high frequency sensors are used to capture the time difference of the electromagnetic wave signal, and the precise location of the partial discharge can be obtained by using triangulation and multi-sensor fusion technology; Among them, it is worth noting that when verifying whether partial discharge actually exists, it is only necessary to obtain the electromagnetic wave signal of any UHF sensor; Specifically, Collect the pulse current signal output by the pulse current sensor, including information such as the amplitude, frequency and phase of the waveform; record the timestamp of the current waveform to facilitate subsequent data synchronization and analysis.

[0031] Collect the electromagnetic wave signals detected by the UHF sensor, including the signal strength, frequency distribution and time characteristics; also need to record the timestamp of the signal to facilitate comparative analysis with the pulse current data.

[0032] Data preprocessing: De-noise, filter and normalize the collected data to reduce the impact of noise and interference; use data smoothing techniques, such as moving average or median filtering, to improve data quality.

[0033] Feature extraction: Extract key features from the processed data. The key features should be able to characterize the characteristics of partial discharge, such as pulse amplitude, width and interval; for ultra-high frequency signals, extract spectrum characteristics, energy distribution and time series characteristics.

[0034] Comparative analysis: The extracted features are compared and analyzed to determine the type and severity of partial discharge; correlation coefficients or other statistical methods are used to evaluate the similarities and differences between two signal features.

[0035] Data fusion: Multiple electromagnetic wave signals and pulse current signals are fused to provide more comprehensive partial discharge information by combining data from different sensors; the relationship between different sensor data is determined through data association analysis, such as correlation coefficient calculation.

[0036] Alarms and reports: Generate alarms based on analysis results and notify maintenance personnel when PD activity exceeding preset thresholds is detected; generate reports regularly to summarize PD activity and provide a basis for maintenance decisions.

[0037] Feedback loop: Establish a feedback loop to adjust and optimize the monitoring system based on actual detection results to improve its accuracy.

[0038] Preferably, it also includes a model building module, which is connected to the data processing module and builds a support vector machine model based on the processed pulse current signal and electromagnetic wave signal.

[0039] In practical applications, pulse current sensors may be affected by various interferences, such as electromagnetic interference, environmental noise, etc.; these interferences may lead to misjudgment or incorrect identification of local discharge signals; therefore, a model needs to be set up to verify whether the pulse current signal is an interference signal; as an embodiment of the present invention, by constructing a support vector machine model, it can help eliminate false alarms caused by environmental noise, electromagnetic interference or other non-discharge activities; in addition, the support vector machine model can help analyze the development trend of local discharge; through long-term monitoring of pulse current signals and electromagnetic wave signals, the development of local discharge can be predicted to provide a basis for maintenance decisions; the model can also predict potential equipment failures and provide a basis for preventive maintenance; by taking measures in advance, more serious equipment damage and power outages can be avoided, including trend analysis of data to monitor the development of local discharge activities and predict potential failure trends. Time series analysis methods, such as autoregressive moving average models, are used to analyze data trends.

[0040] For example, after capturing pulse current signals and electromagnetic wave signals, the feedback of the pulse current signals and electromagnetic wave signals can be substituted into the support vector machine model to predict the development trend of local discharge, and maintenance personnel can be notified in time to perform maintenance to avoid the aggravation of local discharge and cause equipment damage and power outages.

[0041] Preferably, the support vector machine model is used to eliminate interference signals by: According to the formula: ,in, represents the support vector machine model; represents the weight vector, Represents the feature vector after kernel function mapping; represents the bias term; Extract features of the electromagnetic wave signal at the current moment and construct a feature vector X; Input the current feature vector X into the support vector machine model: like If it is a positive number, it is judged that there is partial discharge at the current moment, otherwise it is judged as an interference signal.

[0042] For nonlinear electromagnetic wave signals, a kernel function, such as radial basis function RBF, is first used to map the electromagnetic wave signal to a high-dimensional space to make it linearly separable; the electromagnetic wave signal monitored in real time is then substituted into the support vector machine model for training. When training the support vector machine model, the values ​​of w and b are adjusted so that the result of substituting the characteristic vector of the local discharge signal into the hyperplane equation is as positive as possible, while the result of substituting the characteristic vector of the interference signal into the hyperplane equation is as negative as possible until the training is completed; in actual application, the characteristic vector X is constructed after the electromagnetic wave signal feature is extracted at the current moment. After the characteristic vector X is input into the support vector machine model, if If it is a positive number, it means that the electromagnetic wave signal at the current moment is verified to be partial discharge, otherwise it is an interference signal. When applied to trend prediction, the future trend is predicted by fitting the historical data of partial discharge, thereby effectively avoiding the aggravation of partial discharge and causing equipment damage and power outages.

[0043] Preferably, the data acquisition module also includes a carbon monoxide sensor for collecting carbon monoxide concentration for verifying partial discharge.

[0044] When local discharge occurs in the ring main unit, it will cause the organic insulating material and gas insulating medium to decompose and react chemically with oxygen to produce carbon monoxide gas. During the local discharge process, high-energy electron collisions will cause the chemical bonds in the insulating material to break, generating new compounds that react with oxygen to form carbon monoxide and carbon dioxide at high temperatures. The installation of a carbon monoxide sensor is also helpful in verifying the existence of local discharge. When the UHF sensor fails due to long-term lack of maintenance, the carbon monoxide sensor can be used as an indirect verification indicator, which is helpful in avoiding the aggravation of local discharge.

[0045] Preferably, the method for calculating the time difference and determining the specific occurrence location of the partial discharge is: Calculate the time difference between the electromagnetic wave signal and the pulse current signal; According to the formula: ,in, represents the time difference; ti and tj represent the capture time of electromagnetic wave signals at different positions respectively; Based on the time difference, triangulation is used to determine the specific location where the partial discharge occurs.

[0046] Calculating the time difference can help accurately locate the location of partial discharge; by combining triangulation, the time difference of electromagnetic wave signals captured by multiple ultra-high frequency sensors can be used to accurately locate the specific location where the partial discharge occurs, with a distance error of centimeters, which improves the accuracy and reliability of partial discharge monitoring; as an embodiment of the present invention, when the time difference After calculation, the distance difference from the partial discharge position to the two sensors can be expressed as ; Assuming that the coordinates of multiple UHF sensors are A(x1,y1,z1), B(x2,y2,z2), C(x3,y3,z3) and the partial discharge position coordinates (x,y,z), according to the relationship between time difference and distance difference, the equation is satisfied:

[0047] Based on the above equation, the exact coordinates (x, y, z) of the partial discharge location are calculated.

[0048] A partial discharge monitoring method combining capacitive coupling and ultra-high frequency detection comprises the following steps: S1: The pulse current sensor collects the pulse current signal; S2: Collect electromagnetic wave signals by ultra-high frequency sensor; S3: Processing the pulse current signal and electromagnetic wave signal with the same timestamp; S4: Compare the key features of the pulse current signal and the electromagnetic wave signal to see if they are consistent; S5: Based on the comparison of key features, determine whether partial discharge actually exists. If so, use ultra-high frequency sensors to locate it.

[0049] Working principle: The pulse current sensor can capture the sharp current pulses generated by partial discharge. After capturing the pulse current signal, it processes the pulse current signal and compares it with the threshold. The threshold is a preset value obtained by experiment. When the value of the pulse current signal is greater than the threshold, it indicates that partial discharge is suspected and the position of the partial discharge cannot be located. The ultra-high frequency sensor verifies whether the partial discharge actually exists and locates it; otherwise it is an interference signal. When the value of the pulse current signal is greater than the threshold and the electromagnetic wave signal captured by the ultra-high frequency sensor verifies that the partial discharge actually exists, the alarm unit outputs an alarm signal and notifies the maintenance personnel to perform maintenance and repair on the ring network cabinet with partial discharge to prevent the partial discharge from worsening, resulting in the insulation material being broken down and causing equipment damage or power outages.

[0050] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A partial discharge monitoring system combining capacitive coupling and ultra-high frequency detection, characterized in that: include: A data acquisition module, including a pulse current sensor, for collecting monitoring data; A data processing and analysis module, used for processing and analyzing pulse current signals; The alarm module outputs alarm signals to the cloud platform based on the analysis results: When the analysis result shows that partial discharge is suspected, an alarm signal is output; The data acquisition module also includes: Ultra-high frequency sensor, used to capture electromagnetic wave signals generated by partial discharge; When the analysis result is suspected partial discharge, the UHF sensor calculates the specific location where the partial discharge occurs based on the time difference generated by the electromagnetic wave signal.

2. A partial discharge monitoring system combining capacitive coupling and ultra-high frequency detection according to claim 1, characterized in that: The method in which the alarm module outputs an alarm signal according to the analysis result is: The data acquisition module collects and processes the pulse current signal from the pulse current sensor; comparing the processed pulse current signal with a threshold value; When the value of the pulse current signal is greater than the threshold, a judgment that a partial discharge is suspected to exist is output; When the value of the pulse current signal is less than or equal to the threshold value, the judgment that a partial discharge is suspected to exist is not output.

3. A partial discharge monitoring system combining capacitive coupling and ultra-high frequency detection according to claim 2, characterized in that: The pulse current signal includes pulse amplitude U, pulse count J and pulse phase M; the threshold is a preset value, according to the formula: ,in, Indicates the value of the pulse current signal; , , are the weights of pulse amplitude U, pulse count J and pulse phase M respectively.

4. The partial discharge monitoring system combining capacitive coupling and ultra-high frequency detection according to claim 3 is characterized in that: Based on the judgment that partial discharge is suspected to exist, the data acquisition module also collects electromagnetic wave signals from the ultra-high frequency sensor, and processes the pulse current signal and the electromagnetic wave signal through the data processing and analysis module, and verifies whether partial discharge exists by extracting the characteristics of the pulse current signal and the electromagnetic wave signal.

5. The partial discharge monitoring system combining capacitive coupling and ultra-high frequency detection according to claim 4 is characterized in that: The method for verifying whether partial discharge exists is: Signal synchronization, obtaining electromagnetic wave signals and pulse current signals with time stamp synchronization; Signal denoising: denoising electromagnetic wave signals; Feature extraction, extracting key features from electromagnetic wave signals, including frequency distribution and energy spectral density; Signal comparison, compare and analyze the extracted key features with the features in the pulse current signal: If similar features exist, it is determined that partial discharge exists; otherwise, it does not exist.

6. The partial discharge monitoring system combining capacitive coupling and ultra-high frequency detection according to claim 5, characterized in that: It also includes a model building module, which is connected to the data processing module and builds a support vector machine model based on the processed pulse current signal and electromagnetic wave signal.

7. The partial discharge monitoring system combining capacitive coupling and ultra-high frequency detection according to claim 6, characterized in that: The support vector machine model is used to eliminate interference signals by: According to the formula: ,in, represents the support vector machine model; represents the weight vector, Represents the feature vector after kernel function mapping; represents the bias term; Extract features of the electromagnetic wave signal at the current moment and construct a feature vector X; Input the current feature vector X into the support vector machine model: like If it is a positive number, it is judged that there is partial discharge at the current moment, otherwise it is judged as an interference signal.

8. The partial discharge monitoring system combining capacitive coupling and ultra-high frequency detection according to claim 7, characterized in that: The data acquisition module also includes a carbon monoxide sensor for collecting carbon monoxide concentration for verifying partial discharge.

9. The partial discharge monitoring system combining capacitive coupling and ultra-high frequency detection according to claim 8, characterized in that: The method to calculate the time difference and determine the specific location of partial discharge is: Calculate the time difference between the electromagnetic wave signal and the pulse current signal; According to the formula: ,in, represents the time difference; ti and tj represent the capture time of electromagnetic wave signals at different positions respectively; Based on the time difference, triangulation is used to determine the specific location where the partial discharge occurs.

10. A partial discharge monitoring method combining capacitive coupling and ultra-high frequency detection, characterized in that: The following steps are involved: S1: The pulse current sensor collects the pulse current signal; S2: Collect electromagnetic wave signals by ultra-high frequency sensor; S3: Processing the pulse current signal and electromagnetic wave signal with the same timestamp; S4: Compare the key features of the pulse current signal and the electromagnetic wave signal to see if they are consistent; S5: Based on the comparison of key features, determine whether partial discharge actually exists. If so, use ultra-high frequency sensors to locate it.

Images