Cable joint partial discharge on-line monitoring method and device

Abstract

The invention relates to a cable joint partial discharge on-line monitoring method and device. The cable joint partial discharge on-line monitoring method comprises the steps of determining a discharge type judgment feature and a discharge position judgment feature of a discharge pulse signal; analyzing the discharge type judgment feature to determine a corresponding discharge type; triggering the corresponding environment sensing element combination to obtain corresponding environment data; determining a corresponding discharge position according to the environment data and the discharge position judgment feature; and the extracted discharge type judgment features are analyzed, the discharge type is accurately classified, and the environmental data is used as a compensation factor to help further optimize the discharge position judgment features. And in combination with the environmental data and the discharge position judgment characteristics, the system adopts a multi-modal data fusion algorithm to accurately calculate the specific position of the discharge source. Through multi-dimensional comprehensive analysis of signals, environmental data and positions, the accuracy and reliability of discharge monitoring are significantly improved.

Description

Technical Field

[0001] The invention relates to the technical field of partial discharge monitoring, and in particular to an online monitoring method and device for partial discharge of a cable joint. Background Art

[0002] At present, cable joint partial discharge monitoring technology is widely used in power systems to monitor the health of cable joints in real time. Traditional partial discharge monitoring methods mainly use the discharge pulse signal detected by the sensor, use simple threshold judgment and statistical analysis methods to classify the discharge, and then evaluate the cable joint status.

[0003] Existing cable joint partial discharge monitoring methods have certain limitations. Traditional methods rely on a single signal detection and classification model, which makes it difficult to accurately distinguish different types of partial discharge signals, especially in the presence of complex noise or environmental interference, which can easily lead to false alarms or missed alarms. In addition, existing monitoring methods are often unable to adjust monitoring strategies according to different discharge types, resulting in limited monitoring accuracy and response speed. At the same time, the precise positioning of the discharge position remains a technical difficulty, and existing methods are usually difficult to effectively combine different signal sources to accurately locate the discharge source. Summary of the invention

[0004] In order to solve the problem that traditional methods rely on a single signal detection and classification model and are difficult to accurately distinguish different types of partial discharge signals, the present application provides a cable joint partial discharge online monitoring method and device.

[0005] A method for online monitoring of partial discharge of a cable joint, comprising: If a discharge pulse signal different from background noise is detected, a discharge type determination feature and a discharge position determination feature of the discharge pulse signal are determined; Based on the established classification and recognition model, the discharge type determination features are analyzed to determine the corresponding discharge type; according to the discharge type, the corresponding environmental sensor element combination is triggered to obtain the corresponding environmental data; Determining a corresponding discharge position according to the environmental data and the discharge position determination feature; A corresponding monitoring result is generated based on the association between the discharge position and the discharge type, and the monitoring result is pushed to a corresponding online terminal.

[0006] By adopting the above technical solution, the present application automatically identifies the effective discharge signal that is different from the background noise by comparing the detected discharge pulse signal with the background noise, and on this basis extracts the discharge type determination features, such as time, frequency, amplitude and phase features, and the discharge position determination features, such as signal arrival time difference, etc. Then, based on the established classification and recognition model, the extracted discharge type determination features are analyzed to accurately classify the type of discharge, such as tip discharge, surface discharge, etc. Next, according to the discharge type, the system intelligently triggers the corresponding combination of environmental sensor elements, such as electromagnetic interference, temperature and humidity and other environmental data collectors, and synchronously obtains relevant environmental data, which are used as compensation factors to help further optimize the discharge position determination features. Combined with these environmental data and discharge position determination features, the system uses a multimodal data fusion algorithm to accurately calculate the specific location of the discharge source. Finally, the system generates monitoring results based on the discharge position and discharge type, and pushes the results to the online terminal through the communication network, so that the operator can view and respond in real time. This method significantly improves the accuracy and reliability of discharge monitoring through multi-dimensional comprehensive analysis of signals, environmental data and locations, while enhancing the system's adaptability to complex environments and interference factors, achieving more accurate and efficient partial discharge monitoring.

[0007] Preferably, if a discharge pulse signal different from background noise is detected, the step of determining the discharge type determination feature and the discharge position determination feature of the discharge pulse signal includes: If a discharge pulse signal different from the background noise is detected, the corresponding sampling period is determined; Based on the channel waveform diagram, determining the time characteristics, frequency characteristics, amplitude characteristics and phase characteristics within the sampling period, and determining the time characteristics, frequency characteristics, amplitude characteristics and phase characteristics as discharge type determination characteristics; Based on the double-terminal positioning measurement method, the corresponding discharge position determination feature is determined by calculating the time difference between the discharge pulse signal reaching different sensors within the sampling period.

[0008] By adopting the above technical solution, by detecting the discharge pulse signal that is different from the background noise and determining the corresponding sampling period, it is possible to ensure that the effective discharge signal is captured within the accurate time window, thereby avoiding misjudgment of irrelevant signals and ensuring the accuracy of the data; by determining the time characteristics, frequency characteristics, amplitude characteristics and phase characteristics based on the channel waveform diagram and integrating them into the discharge type determination characteristics, it is possible to comprehensively extract features based on multi-dimensional information, thereby improving the recognition accuracy of the discharge type; by calculating the time difference between the discharge pulse signal arriving at different sensors based on the double-end positioning measurement method, the position of the discharge source can be accurately determined, thereby improving the accuracy and reliability of the discharge position determination.

[0009] Preferably, the step of analyzing the discharge type determination feature based on the established classification recognition model to determine the corresponding discharge type includes: Matching the type weight coefficient corresponding to each of the discharge type determination features, and generating a corresponding type feature vector according to the product of the type weight coefficient and the discharge type determination feature; Adding each of the type of feature vectors to generate a corresponding comprehensive feature vector; Substitute the comprehensive feature vector into the established classification and recognition model to output the corresponding discharge type.

[0010] By adopting the above technical solution, by matching the type weight coefficient corresponding to each discharge type determination feature, each feature can be weighted according to its importance, ensuring that the model can more accurately capture the essential characteristics of the discharge type, thereby improving the classification accuracy; by adding the type feature vectors to generate a comprehensive feature vector, multiple features can be fused to enhance the recognition ability of the model and reduce interference between features; by substituting the comprehensive feature vector into the classification and recognition model, the discharge type can be quickly derived, thereby realizing automated and efficient fault diagnosis.

[0011] Preferably, the step of triggering a corresponding combination of environmental sensor elements to acquire corresponding environmental data according to the discharge type includes: Determining corresponding associated dimensions according to the discharge type, and determining a triggering order of each associated dimension; The corresponding sensor elements are matched according to the associated dimensions, and the various sensor elements are integrated into a corresponding environmental sensor element combination; based on the trigger order, the corresponding sampling interval is determined within the sampling period, and according to the trigger order and the sampling interval, the corresponding environmental sensor element combination is triggered to obtain the corresponding environmental data.

[0012] By adopting the above technical solution, by determining the corresponding correlation dimensions according to the discharge type and determining the triggering order of each correlation dimension, the collection priority of the sensor can be dynamically adjusted to ensure that the most relevant environmental data is obtained under different discharge types, thereby improving the efficiency and pertinence of data collection; by triggering the corresponding combination of environmental sensor elements to obtain environmental data according to the triggering order and sampling interval, it can be ensured that the environmental data is collected at the most appropriate time, thereby improving the response speed and accuracy of the system.

[0013] Preferably, the step of determining the corresponding discharge position according to the environmental data and the discharge position determination feature includes: Preprocessing the environmental data and the discharge position determination features; Based on regression analysis, determine the compensation factor i corresponding to the environmental data i, wherein the environmental data i is the i-th environmental data, and the compensation factor i is the influence coefficient of the i-th environmental data on the discharge position; Determining an initial discharge position S1 according to the discharge position determination feature; The environmental data i, compensation factor i and initial discharge position S1 are substituted into the established compensation formula to generate a compensated discharge position S2. The established compensation formula is: S2=S1+∑(compensation factor i×environmental data i).

[0014] By adopting the above technical scheme, by preprocessing the environmental data and the discharge position determination features, it is possible to eliminate noise and unify the data scale, thereby ensuring the accuracy of subsequent analysis; by determining the compensation factor corresponding to the environmental data based on regression analysis, it is possible to quantify the impact of environmental factors on the discharge position, thereby providing an accurate adjustment basis for position correction; by substituting the environmental data, the compensation factor and the initial discharge position into the compensation formula, the discharge position can be accurately compensated, thereby improving the accuracy of the discharge position determination.

[0015] Preferably, in the step of generating a corresponding monitoring result based on the association between the discharge position and the discharge type, the monitoring result at least includes an alarm instruction and a visualization model, and the step further includes: According to the discharge position, determining a corresponding associated impact area; Determining, according to the discharge type, an association result of the association impact area; Determine whether the correlation result exceeds a preset result threshold, if not, continue to determine, if exceeded, generate a corresponding alarm instruction; A corresponding visualization model is established according to the association impact area and the association result.

[0016] By adopting the above technical solution, by determining the corresponding associated impact area according to the discharge position, the potential impact area related to the discharge source can be identified, thereby providing comprehensive information for fault analysis; by determining the associated results of the associated impact area according to the discharge type, the discharge type can be matched with the possible fault area, thereby improving the accuracy of fault location; by judging whether the associated result exceeds the preset threshold, if exceeded, an alarm instruction is generated, and an alarm can be issued in time in the case of abnormal discharge to ensure the safety of the system; by establishing a visualization model based on the associated impact area and the associated results, the monitoring results can be intuitively displayed to help operators make decisions quickly.

[0017] A cable joint partial discharge online monitoring device uses a cable joint partial discharge online monitoring method. The cable joint partial discharge online monitoring device comprises a mutual inductance acquisition component, a cable joint, a partial discharge collector, an environmental data collector and a monitoring host. The cable joint is sleeved on the cable and connected to the cross-connected grounding wire. The signal output end of the mutual inductance acquisition component is connected to the signal input end of the partial discharge collector. The signal output end of the partial discharge collector is connected to the first signal input end of the monitoring host. The signal output end of the environmental data collector is connected to the second signal input end of the monitoring host.

[0018] By adopting the above technical solutions, efficient data collection, analysis and processing can be achieved in the online monitoring device for partial discharge of cable joints, thereby improving the accuracy, real-time and intelligence level of the monitoring system; by integrating mutual inductance acquisition components, partial discharge collectors, environmental data collectors and monitoring hosts, multi-sensor collaborative work can be achieved, and partial discharge signals of cable joints can be effectively captured and analyzed, providing comprehensive support for fault diagnosis.

[0019] Preferably, the mutual inductance collection component includes a high-frequency pulse current transformer, the signal collection end of the high-frequency pulse current transformer is coaxially spaced with the cross-interconnected grounding wire, and the signal output end of the high-frequency pulse current transformer is connected to the signal input end of the partial discharge collector.

[0020] By adopting the above technical solution, by applying the high-frequency pulse current transformer to the mutual inductance acquisition component, the high-frequency discharge signal of the cable joint can be accurately captured, ensuring the sensitivity and real-time performance of the monitoring system to discharge events; by connecting the signal output of the mutual inductance acquisition component with the local discharge collector, the signal data can be efficiently transmitted, ensuring the stability and reliability of data acquisition.

[0021] Preferably, the mutual inductance collection component further comprises a power frequency phase mutual inductor, the signal collection end of the power frequency phase mutual inductor is coaxially spaced from the cable, and the signal output end of the power frequency phase mutual inductor is connected to the signal input end of the partial discharge collector.

[0022] By adopting the above technical solution and applying the power frequency phase mutual inductor to the mutual inductance acquisition component, the power frequency signal of the cable can be monitored, providing more auxiliary information for the accurate identification and classification of the discharge signal; by connecting with the local discharge collector, the precision and accuracy of signal acquisition can be further improved, ensuring the accurate determination of the discharge type and location.

[0023] Preferably, the monitoring host is a FPGA processor.

[0024] By adopting the above technical solution and setting the monitoring host as an FPGA processor, high-speed data acquisition and processing can be achieved, ensuring a rapid response to cable joint discharge events; the parallel processing capability of FPGA can improve the system processing capability, achieve efficient processing of large-scale data, and ensure the real-time and stability of the system.

[0025] In summary, the present application includes at least one of the following beneficial technical effects: This application automatically identifies the effective discharge signal that is different from the background noise by comparing the detected discharge pulse signal with the background noise, and on this basis extracts the discharge type determination features, such as time, frequency, amplitude and phase features, and discharge position determination features, such as signal arrival time difference, etc. Then, based on the established classification and recognition model, the extracted discharge type determination features are analyzed to accurately classify the type of discharge, such as tip discharge, surface discharge, etc. Next, according to the discharge type, the system intelligently triggers the corresponding combination of environmental sensor elements, such as electromagnetic interference, temperature and humidity and other environmental data collectors, and simultaneously obtains relevant environmental data, which are used as compensation factors to help further optimize the discharge position determination features. Combined with these environmental data and discharge position determination features, the system uses a multimodal data fusion algorithm to accurately calculate the specific location of the discharge source. Finally, the system generates monitoring results based on the discharge location and discharge type, and pushes the results to the online terminal through the communication network, so that the operator can view and respond in real time. This method significantly improves the accuracy and reliability of discharge monitoring through multi-dimensional comprehensive analysis of signals, environmental data and locations, while enhancing the system's adaptability to complex environments and interference factors, achieving more accurate and efficient partial discharge monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flow chart of a method for online monitoring of partial discharge of cable joints in one embodiment of the present application.

[0027] Figure 2 This is a flowchart for implementing step S10 in a method for online monitoring of partial discharge of a cable joint in one embodiment of the present application; Figure 3 This is a flowchart for implementing step S20 in a method for online monitoring of partial discharge of a cable joint in one embodiment of the present application; Figure 4 This is a flowchart for implementing step S30 in a method for online monitoring of partial discharge of a cable joint in an embodiment of the present application; Figure 5 This is a flowchart for implementing step S40 in a method for online monitoring of partial discharge of a cable joint in one embodiment of the present application; Figure 6This is a flowchart for implementing step S50 in a method for online monitoring of partial discharge of a cable joint in one embodiment of the present application; Figure 7 It is a specific structural schematic diagram of an online monitoring device for partial discharge of a cable joint in one embodiment of the present application. DETAILED DESCRIPTION

[0028] The present application is further described in detail below in conjunction with the accompanying drawings.

[0029] In one embodiment, if Figure 1 As shown, the present application discloses an online monitoring method for partial discharge of a cable joint, which specifically includes the following steps: S10, if a discharge pulse signal different from the background noise is detected, determining a discharge type determination feature and a discharge position determination feature of the discharge pulse signal; In this embodiment, if a discharge pulse signal that is different from the background noise is detected, the system filters and eliminates the noise of the input signal through a precise signal processing method to ensure that only valid discharge signals are retained. In this process, the system first uses a bandpass filter to filter out low-frequency and high-frequency noise in the signal to ensure that the signal is concentrated in the frequency band of local discharge. Then, the system denoises the signal and uses noise estimation and adaptive filtering technology to separate the real discharge signal from the environmental noise. Through this series of processing, the signal-to-noise ratio of the signal can be effectively improved, ensuring that subsequent analysis only relies on valid discharge signals, thereby improving the accuracy and stability of discharge signal recognition.

[0030] S20, analyzing the discharge type determination features based on the established classification and recognition model to determine the corresponding discharge type; In this embodiment, the discharge type determination features are analyzed based on the established classification recognition model. By using advanced machine learning models, such as support vector machines (SVM), decision trees, neural networks, etc., the system can input the extracted signal features such as time, frequency, amplitude, phase, etc. into the classification model, thereby automatically identifying the type of discharge. The process first trains the model through a training data set to ensure that it can accurately determine the discharge type based on specific discharge characteristics. Through the judgment of the classification model, the discharge type (such as tip discharge, surface discharge or air gap discharge, etc.) can be quickly identified, thereby providing a basis for subsequent fault diagnosis and location positioning. Since the model is trained based on a large amount of real data, it can effectively improve the accuracy and robustness of discharge type classification.

[0031] S30, triggering a corresponding combination of environmental sensor elements to obtain corresponding environmental data according to the discharge type; In this embodiment, according to the type of discharge, the system triggers the corresponding combination of environmental sensing elements to obtain environmental data. According to different types of discharge, the system automatically selects appropriate environmental sensors for data collection. For example, when a tip discharge is detected, the system will give priority to starting the electromagnetic field strength sensor, because the tip discharge is usually accompanied by strong electromagnetic interference; and when a surface discharge is detected, the temperature and humidity sensors will be activated, because surface discharge is often closely related to temperature and humidity changes. After triggering the corresponding combination of environmental sensing elements, the system can collect accurate environmental data such as electromagnetic field strength, temperature, humidity, etc. in real time, and provide rich environmental information to assist subsequent analysis. This process enables the monitoring system to dynamically adjust the data collection strategy for different discharge types, thereby improving the accuracy and adaptability of monitoring.

[0032] S40, determining a corresponding discharge position according to the environmental data and the discharge position determination feature; In this embodiment, the system can determine the accurate discharge position by combining environmental data and discharge position determination features. By fusing environmental data (such as temperature and humidity, electromagnetic field, etc.) with the time difference, signal strength and other characteristics of the discharge position, the system can accurately locate based on multimodal data. Using positioning algorithms such as double-end positioning and Kalman filtering, combined with the arrival time difference of the discharge signal, the system can determine the specific location of the discharge source. In addition, environmental data is used as a compensation factor, and regression analysis can be used to correct the discharge signal deviation caused by environmental factors (such as humidity changes, electromagnetic interference, etc.), making the position determination more accurate. This compensation process effectively eliminates the negative impact of environmental interference on position determination and ensures more accurate discharge position positioning.

[0033] S50: Generate corresponding monitoring results based on the association between the discharge position and the discharge type, and push the monitoring results to the corresponding online terminal.

[0034] In this embodiment, the corresponding monitoring results are generated based on the association between the discharge position and the discharge type, and the monitoring results are pushed to the corresponding online terminal. First, the system combines the discharge type, location and environmental data to generate a comprehensive monitoring report. These reports not only include the discharge type (such as tip discharge, surface discharge, etc.), but also provide the precise location of the discharge source and the associated information of the potential fault area. The monitoring results can be displayed through a graphical interface, such as using a thermal map or a three-dimensional model to display the location of the discharge source, to help operators intuitively understand the status of the cable joint. At the same time, the monitoring results will be pushed to the operator's online terminal (such as PC, mobile phone, etc.) in real time through a communication protocol (such as WebSocket, MQTT, etc.) to ensure that the operator can quickly obtain fault information and respond. The system can automatically generate an alarm when the monitoring results exceed the set threshold and provide real-time feedback, thereby greatly improving the response speed and processing efficiency of cable joint faults.

[0035] In one embodiment, if Figure 2 As shown, in step S10, that is, if a discharge pulse signal different from background noise is detected, the step of determining the discharge type determination feature and the discharge position determination feature of the discharge pulse signal includes: S101, if a discharge pulse signal different from background noise is detected, determine a corresponding sampling period; In this embodiment, if a discharge pulse signal that is different from the background noise is detected, the corresponding sampling period is determined. The purpose of this step is to determine which signals are true local discharge signals rather than background noise or other irrelevant signals through signal analysis technology. Through the strength and frequency analysis of real-time signals, the system can identify meaningful discharge pulses and set appropriate sampling periods to ensure that the details of the entire discharge process can be fully captured during the sampling period. The setting of the sampling period is based on the duration, frequency range and sampling rate requirements of the discharge signal, ensuring that the discharge signal can be fully collected within a certain period of time without missing important discharge information due to a too short sampling period. Specifically, when it is detected that the signal strength reaches a predetermined threshold, the system will automatically set an adaptive sampling period that is long enough to cover the complete discharge signal to avoid affecting the collection effect due to environmental noise or interference signals.

[0036] S102, based on the channel waveform diagram, determining the time characteristics, frequency characteristics, amplitude characteristics and phase characteristics within the sampling period, and determining the time characteristics, frequency characteristics, amplitude characteristics and phase characteristics as discharge type determination characteristics; In this embodiment, based on the channel waveform diagram, the time characteristics, frequency characteristics, amplitude characteristics and phase characteristics within the sampling period are determined, and the time characteristics, frequency characteristics, amplitude characteristics and phase characteristics are all determined as discharge type determination characteristics. In this step, the system extracts four main characteristics of the waveform signal collected by the sensor by analyzing the waveform signal: time characteristics, frequency characteristics, amplitude characteristics and phase characteristics. Specifically, the time characteristics refer to the duration of the signal, the pulse interval and the start and end time of the signal, which helps to judge the periodicity and stability of the discharge; the frequency characteristics are obtained by Fourier transforming the signal to obtain the spectrum distribution of the signal, which reflects the frequency range of the discharge. The tip discharge usually appears as a high-frequency signal, while the surface discharge may appear as a lower frequency signal; the amplitude characteristics are used to measure the intensity of the signal and show the amplitude of the discharge. Different types of discharge signals may have significant differences in intensity. Signals with larger amplitudes may represent serious faults; the phase characteristics describe the phase changes of the signal within the time period. The tip discharge usually occurs at a specific stage of the power grid cycle and has significant phase characteristics. These features are extracted through mathematical models and used for subsequent discharge type determination, which can improve the accuracy of discharge type identification and help distinguish different discharge modes.

[0037] S103 , based on the double-end positioning measurement method, by calculating the time difference between the discharge pulse signal reaching different sensors within the sampling period, the corresponding discharge position determination feature is determined.

[0038] In this embodiment, based on the double-end positioning measurement method, the corresponding discharge position determination feature is determined by calculating the time difference of the discharge pulse signal arriving at different sensors within the sampling period. By installing sensors at both ends or multiple positions of the cable, the system can capture the time difference of the discharge signal propagation. Specifically, the system uses the double-end positioning measurement method, that is, by measuring the propagation time difference of the discharge signal from the discharge source to different sensors, to calculate the position of the discharge source. Since the propagation speed of the electromagnetic signal in different media is known, by accurately measuring the time difference of the signal arriving at each sensor, the system can accurately calculate the coordinates or area of ​​the discharge source. The time difference of the discharge signal arrival is proportional to the distance. Using this principle, the system can calculate the distance between the discharge source and the sensor through an algorithm, and finally determine the exact position of the discharge source. In addition, the system can also compensate for the time difference through environmental data (such as temperature and humidity, electromagnetic field interference, etc.) to further improve the positioning accuracy.

[0039] In one embodiment, if Figure 3 As shown, in step S20, that is, based on the established classification and recognition model, the discharge type determination feature is analyzed to determine the corresponding discharge type, including: S201, matching the type weight coefficient corresponding to each discharge type determination feature, and generating a corresponding type feature vector according to the product of the type weight coefficient and the discharge type determination feature; In this embodiment, the type weight coefficient corresponding to each discharge type determination feature is matched, and the corresponding type feature vector is generated according to the product of the type weight coefficient and the discharge type determination feature. Specifically, the discharge type determination feature is a feature extracted from multiple dimensions by analyzing the discharge pulse signal captured by the sensor, such as the time feature (pulse width, pulse interval), frequency feature (spectrum, frequency distribution of the signal), amplitude feature (signal strength, amplitude change) and phase feature (phase change of the signal at different time points). These features reflect different aspects of the discharge signal and can provide important information about the discharge type. In order to improve the classification accuracy, the system assigns a weight coefficient to each feature, and the values ​​of these weight coefficients reflect the contribution of different features to the discharge type determination. The weight coefficient is obtained based on historical data or through machine learning training to ensure that the contribution of each feature matches its actual impact on the discharge type determination. For example, for tip discharge, the frequency feature may have a higher weight because the tip discharge is usually manifested as a high-frequency signal, while for surface discharge, the time feature and amplitude feature may be more important because these features can better capture the signal characteristics of surface discharge. By multiplying the weight coefficient with each feature, the system generates weighted feature values, which constitute a type feature vector representing the comprehensive characteristics of the discharge signal.

[0040] S202, adding each type of feature vector to generate a corresponding comprehensive feature vector; In this embodiment, each type of feature vector is added to generate a corresponding comprehensive feature vector. Specifically, the system merges the weighted results from each different discharge type determination feature. Each feature vector represents the information of the discharge signal in a certain dimension. For example, the time feature vector represents the time attribute of the signal, the frequency feature vector represents the frequency distribution of the signal, and the amplitude feature vector represents the intensity change of the signal. By adding these feature vectors, the system can obtain a comprehensive feature vector that combines all the important features of the discharge signal. This comprehensive feature vector can provide a comprehensive description of the discharge signal, integrate the influence of each feature, and thus improve the accuracy of classification. For example, the time, frequency, amplitude and phase features may describe the type of discharge from different angles, and after weighted fusion, they can more comprehensively reflect the actual situation of the discharge signal. Through this fusion, the system can simplify complex signal features into a unified vector that can be used for classification, thereby improving the determination accuracy of the model.

[0041] S203, substituting the comprehensive feature vector into the established classification and recognition model, and outputting the corresponding discharge type.

[0042] In this embodiment, the comprehensive feature vector is substituted into the established classification recognition model, and the corresponding discharge type is output. The system analyzes the generated comprehensive feature vector by inputting it into a pre-trained classification model, which can be a support vector machine (SVM), a decision tree, a random forest, a deep neural network (DNN), etc. The classification model outputs the corresponding discharge type, such as tip discharge, surface discharge, or air gap discharge, by calculating the degree of match between the comprehensive feature vector and the pattern learned internally. The key to this process is that the classification model can quickly and accurately determine the type of discharge signal based on the information in the comprehensive feature vector. During model training, the system uses a large number of labeled data sets, and by continuously adjusting internal parameters, the model can accurately classify new discharge signals in practical applications. In the process of outputting the discharge type, the model weights each feature according to the feature weights of different discharge types, so as to obtain the most appropriate classification result, thereby improving the accuracy of discharge type recognition and the reliability of the system.

[0043] Through the above steps, the system can accurately determine the discharge type based on the real-time collected signal data and the corresponding feature vectors, providing strong data support for subsequent discharge location positioning and environmental impact analysis.

[0044] In one embodiment, if Figure 4 As shown, in step S30, i.e., the step of triggering the corresponding environmental sensor element combination to obtain the corresponding environmental data according to the discharge type, it includes: S301, determining corresponding associated dimensions according to the discharge type, and determining a triggering order of each associated dimension; In this embodiment, according to the discharge type, the system first determines the corresponding associated dimensions and determines the triggering order of each associated dimension. The discharge type (such as tip discharge, surface discharge, etc.) has different effects on environmental conditions, so the system identifies the environmental characteristic dimensions related to it according to different types of discharge signals. The associated dimensions refer to environmental factors that affect the discharge type, such as temperature, humidity, electromagnetic field strength, vibration, etc. Each discharge type may be closely related to a specific environmental dimension. For example, for tip discharge, electromagnetic field strength may be the most important associated dimension, while for surface discharge, temperature and humidity may be more critical. By selecting appropriate associated dimensions according to the discharge type, the system can ensure that only environmental sensors related to the discharge type are triggered for data collection, thereby improving the accuracy and efficiency of data collection. In addition, the system determines their triggering order according to the characteristics of these associated dimensions, ensuring that the collected environmental data can be synchronized with the characteristics of the discharge signal, further improving the timeliness of monitoring.

[0045] S302, matching corresponding sensor elements according to the associated dimensions, and integrating the sensor elements into a corresponding environmental sensor element combination; In this embodiment, the corresponding sensor elements are matched according to the associated dimensions, and the various sensor elements are integrated into the corresponding environmental sensor element combination. Different discharge types require different environmental data to compensate and support judgment. Therefore, the system selects the most suitable sensor element for each dimension according to the determined associated dimensions. For example, if the associated dimensions are temperature and humidity, the system may select a temperature and humidity sensor; if it is electromagnetic field strength, the system may select an electromagnetic field sensor. After selecting the sensor, the system integrates these sensor elements to form a sensor combination to efficiently collect multi-dimensional environmental data. The integrated sensor element combination may include multiple sensor modules, which can simultaneously obtain environmental data from multiple dimensions and provide sufficient information for accurate judgment of the discharge type. In this way, the system can dynamically select and activate sensors according to the actual discharge type requirements, thereby avoiding unnecessary sensor waste and improving the resource utilization and acquisition efficiency of the system.

[0046] S303: Based on the triggering order, determine the corresponding sampling interval within the sampling period, and trigger the corresponding combination of environmental sensor elements to obtain corresponding environmental data according to the triggering order and the sampling interval.

[0047] In this embodiment, based on the trigger order, the corresponding sampling interval is determined within the sampling period, and according to the trigger order and the sampling interval, the corresponding environmental sensor element combination is triggered to obtain the corresponding environmental data. The system determines the start time of different sensor element combinations according to the aforementioned trigger order to ensure that different environmental data are collected at the best time. The design of the trigger order takes into account the characteristics of different types of discharge signals. For example, under the influence of the tip discharge signal, the electromagnetic field sensor may need to be triggered more frequently, while the trigger frequency of the temperature and humidity sensor is lower. The system controls the frequency of data collection by determining the sampling interval (i.e., the time interval for the sensor to start) to ensure the accuracy and timeliness of data collection. The sampling interval will be dynamically adjusted according to the duration of the discharge signal, the speed of environmental changes, and the monitoring requirements to avoid excessive or missed environmental data. In this way, the system can not only improve the efficiency of data collection, but also ensure that the collected data can accurately reflect the changes in the environment, and further optimize the judgment and analysis of the discharge type.

[0048] Specifically, assume that the system detects a tip discharge signal. Based on the characteristics of the tip discharge, the system determines the environmental dimensions associated with it, such as electromagnetic field strength and temperature and humidity. These two dimensions are crucial to the monitoring of tip discharge, because tip discharge is usually accompanied by strong electromagnetic field changes, and changes in environmental temperature and humidity may affect the intensity and propagation characteristics of the discharge. The system selects these two dimensions as relevant features according to the type of discharge, and sets their triggering order based on experience or preset rules. First, the system determines that the electromagnetic field strength is the most critical associated dimension, so the electromagnetic field sensor is triggered first. In order to ensure that the electromagnetic field data can be obtained in time during the discharge process, the system sets a short time interval to quickly collect electromagnetic field data, while the triggering of the temperature and humidity sensor is relatively slow, and the sampling interval is slightly longer. Through this sampling strategy, the system enables the electromagnetic field data to keep up with the changes in the discharge event, while the temperature and humidity data are collected relatively smoothly to avoid excessive interference with data collection. The system then starts the electromagnetic field strength sensor and collects data according to these triggering orders. Assuming that during the discharge process, the fluctuation of the electromagnetic field intensity reaches its maximum near the signal peak, the system captures these changing electromagnetic field data and combines them with real-time environmental data (such as temperature and humidity changes) to generate an environmental data combination. Based on the timing of the discharge signal and these environmental data, the system finally completes the confirmation of the tip discharge type and accurately calculates the location of the discharge source. Through the precise control of the trigger sequence and sampling interval, the system effectively ensures the timely collection of key data while avoiding excessive irrelevant data collection, further improving the accuracy and efficiency of monitoring. For example, the real-time change of the electromagnetic field intensity is crucial to the judgment of the tip discharge, while the change of temperature and humidity provides supplementary information on the long-term impact of the discharge. This mechanism of dynamically adjusting the triggering of the environmental sensor effectively improves the system's adaptability under different discharge types, thereby providing more accurate and reliable partial discharge monitoring in practical applications.

[0049] In one embodiment, if Figure 5 As shown, in step S40, i.e., the step of determining the corresponding discharge position according to the environmental data and the discharge position determination feature, includes: S401, preprocessing environmental data and discharge position determination features; In this embodiment, the environmental data and the discharge position determination features are preprocessed. The preprocessing step is a process of cleaning and optimizing the original environmental data and discharge position features collected from the sensor to ensure that their quality meets the requirements of subsequent analysis. Since the environmental data and the discharge position determination features may be affected by signal noise, external interference and equipment errors, filtering, denoising, standardization and outlier detection are required. For example, environmental data such as temperature, humidity, electromagnetic field and other sensors may produce small errors due to environmental fluctuations, or data drift caused by sensor aging; the discharge position determination features may cause signal delay or distortion due to signal interference. By adopting data smoothing techniques such as low-pass filters, median filters, and sliding average methods, the system can effectively remove high-frequency noise and instantaneous outliers; in addition, through standardization processing (such as scaling the data to a uniform range), different types of environmental data can be processed on the same scale. This preprocessing process provides high-quality input data for subsequent compensation analysis, regression calculations and precise positioning of the discharge position, thereby improving the accuracy and stability of the entire system.

[0050] S402, determining a compensation factor i corresponding to environmental data i based on regression analysis, wherein environmental data i is the i-th environmental data, and compensation factor i is an influence coefficient of the i-th environmental data on the discharge position; In this embodiment, based on regression analysis, the compensation factor i corresponding to the environmental data i is determined, wherein the environmental data i is the i-th environmental data, and the compensation factor i is the influence coefficient of the i-th environmental data on the discharge position. The core of this step is to quantify the influence of different environmental factors on the discharge position through a regression analysis model. The regression analysis model is established based on historical data or experimental data, which contains the relationship between multiple environmental variables and the discharge position. Specifically, the system finds out which environmental data (such as temperature, humidity, electromagnetic field strength, etc.) has a strong correlation with the offset of the discharge position by performing regression analysis on the existing environmental data and the actual discharge position data. Taking temperature as an example, the insulating material of the cable may expand or contract at different temperatures, resulting in changes in the speed of signal propagation, thereby affecting the accuracy of the discharge position. Through regression analysis, the system can assign a compensation factor to each environmental data, and this compensation factor reflects the degree of influence of the environmental data on the discharge position. For example, the influence factor of temperature on the discharge position may be higher, while the influence of humidity may be lower. This compensation factor can dynamically adjust and compensate for the errors caused by environmental changes to ensure that the calculation of the discharge position is more accurate.

[0051] S403, determining an initial discharge position S1 according to the discharge position determination feature; In this embodiment, the initial discharge position S1 is determined according to the discharge position determination feature. The discharge position determination feature is usually a preliminary position calculated by measuring the time difference of the signal reaching different sensors (for example, by the double-end positioning method). Specifically, the system estimates the position of the discharge source based on the time difference of signal propagation. The time difference of the signal reaching the sensor reflects the relative distance of the discharge source from each sensor, and these time differences will be used to infer the preliminary value of the discharge position through triangulation or multi-point positioning algorithm. Due to the influence of factors such as environmental interference and sensor error, this initial discharge position may not be completely accurate, but it provides a basis for subsequent compensation steps. For example, assuming that the arrival times of the signals detected by two sensors are T1 and T2 respectively, the system will calculate the distance difference from the discharge source to the two sensors based on the known signal propagation speed, thereby obtaining a preliminary discharge position S1. This is a preliminary estimated position, and the accuracy still needs to be improved through compensation of environmental data.

[0052] S404, substituting the environmental data i, the compensation factor i and the initial discharge position S1 into the established compensation formula to generate the compensated discharge position S2, the established compensation formula is: S2 = S1 + ∑ (compensation factor i × environmental data i).

[0053] In this embodiment, the environmental data i, the compensation factor i and the initial discharge position S1 are substituted into the established compensation formula to generate the compensated discharge position S2. The core of this step is to correct the initial discharge position through the compensation formula, taking into account the influence of environmental factors on the position determination. The compensation formula is a mathematical model, which is usually obtained through regression analysis. It combines environmental data (such as temperature, humidity, electromagnetic field strength, etc.) and the corresponding compensation factor with the initial discharge position to calculate the final discharge position. For example, it is assumed that the initial discharge position S1 is calculated based on the time difference, but in a high temperature environment, the insulation material of the cable may expand, resulting in a change in the signal propagation speed, thereby causing the initial position S1 to deviate. The system determines the influence coefficient of temperature on the position through regression analysis, and substitutes the compensation factor and the temperature data into the compensation formula to correct the initial position S1 and obtain the compensated discharge position S2. This compensated discharge position S2 can effectively eliminate the influence of environmental factors on the initial position and improve the accuracy of the discharge position. Through this compensation mechanism, the system can maintain high-precision discharge position determination under different environmental conditions.

[0054] In one embodiment, if Figure 6 As shown, in step S50, i.e., the step of generating corresponding monitoring results based on the association between the discharge position and the discharge type, the monitoring results at least include an alarm instruction and a visualization model, and the step further includes: S501, determining a corresponding associated impact area according to a discharge position; In this embodiment, the corresponding associated impact area is determined according to the discharge location. This step is performed by accurately calculating the location of the discharge source and determining the area that may be affected according to the layout of the equipment and the structure of the power system. Specifically, the system uses the discharge location data and combines the geometric structure of the cable, the device connection method, and the layout of the electrical equipment to infer the impact area that the discharge may affect. For example, if the discharge occurs near the cable joint, the system may define a circular area centered on the joint as the impact range. If other cables or electrical equipment around the equipment may also be affected by the discharge, the system includes these areas. To achieve this, the system first calculates the distance between each device or area and the discharge source and the possible impact range based on the known relative position of the discharge source location and other equipment. The size and range of the impact area will vary depending on the intensity, propagation mode and spatial distribution of the discharge type. The stronger the discharge type, the larger the impact range. In this way, the system can accurately identify the area affected by the discharge, thereby providing a specific basis for subsequent alarm generation and maintenance decisions.

[0055] S502, determining the association result of the association impact area according to the discharge type; In this embodiment, the association result of the associated impact area is determined according to the discharge type. This step is calculated according to the actual impact of different discharge types on equipment and areas. Different types of discharges (such as tip discharge, surface discharge, etc.) have different impact characteristics on the environment and equipment. The system will further analyze the impact area according to the characteristics of the discharge type. Tip discharge is usually accompanied by strong electromagnetic interference, which may extend to a large area near the cable, while surface discharge usually affects a smaller range and may be limited to the surface of the cable. By combining the amplitude, frequency and phase characteristics of the discharge signal, the system can evaluate the specific impact of the discharge type on the impact area. For example, if the electromagnetic field strength caused by the tip discharge exceeds the safety standard, the system will calculate the potential impact of the electromagnetic interference on the equipment, thereby determining the association result of the area. Based on these calculations, the system can generate an impact result for each associated area, which reflects the impact of the discharge type on each area and helps operators assess the potential risk of failure.

[0056] S503, determining whether the correlation result exceeds a preset result threshold, if not, continuing the determination, if exceeding, generating a corresponding alarm instruction; In this embodiment, it is determined whether the association result exceeds the preset result threshold. If not, the determination is continued. If it exceeds, a corresponding alarm instruction is generated. The purpose of this step is to determine whether emergency action is needed based on the calculation of the discharge type and the impact result. The system first evaluates the threat of the current discharge to the device by comparing the association result with the set safety threshold. If the association result does not exceed the preset threshold, the system will continue to monitor the cable and device status and maintain real-time detection of environmental data; if it exceeds the threshold, the system will consider that the discharge has exceeded the safety range and there is a risk of potential equipment damage or failure, thereby triggering an alarm instruction. The alarm instruction can be a flashing warning sign on the system interface, a sound alarm, or sent to the operator through SMS, email, etc. This alarm mechanism can help the operator understand the equipment status in a timely manner and take measures to prevent the expansion of the fault or damage to the equipment. For example, if the calculation results of the discharge location and electromagnetic interference indicate that the equipment may be seriously affected, the system will automatically trigger an alarm to prompt the operator to check the health of the cable connector or surrounding equipment.

[0057] S504: Establish a corresponding visualization model according to the associated impact area and the associated result.

[0058] In this embodiment, a corresponding visualization model is established according to the associated impact area and the associated results. The core of this step is to display the above calculation results and associated data in a graphical way, so that the operator can intuitively understand the impact range and risk level of the discharge. The system integrates the location information of the discharge source, the discharge type, the impact area and the relevant calculation results, and displays them through visualization tools. Specific visualization forms may include heat maps, two-dimensional or three-dimensional models, etc., which illustrate the location of the discharge source and the impact of the discharge on the equipment and area. For example, the system can use a heat map to display the electromagnetic interference intensity near the cable joint, with the area with higher intensity marked in red and the area with weaker intensity marked in green; or use a three-dimensional map to mark the spatial position of the cable joint and other equipment, and display the location and impact area of ​​the discharge source through markings of different colors or intensities. Through this visualization display, the operator can quickly identify the fault area that needs to be handled first, and improve the fault response efficiency and decision-making ability. In addition, the system can also generate detailed reports or data charts for each impact area to help managers perform subsequent maintenance and analysis.

[0059] like Figure 7As shown, a cable joint partial discharge online monitoring device uses a cable joint partial discharge online monitoring method. A cable joint partial discharge online monitoring device includes a mutual inductance acquisition component, a cable joint 1, a partial discharge collector 2, an environmental data collector 3 and a monitoring host 4. The cable joint 1 is sleeved on the cable and connected to the cross-interconnected grounding wire. The signal output end of the mutual inductance acquisition component is connected to the signal input end of the partial discharge collector 2, the signal output end of the partial discharge collector 2 is connected to the first signal input end of the monitoring host 4, and the signal output end of the environmental data collector 3 is connected to the second signal input end of the monitoring host 4.

[0060] In this embodiment, the cable joint 1 is one of the key components of the discharge monitoring system, which is directly installed at the joint of the cable. The main function of the cable joint 1 is to provide the source of the discharge signal for the entire monitoring system. The cable joint 1 is connected to the cross-connected grounding wire to ensure the stability of the electrical connection of the cable joint 1 and to ensure that the local discharge signal at the cable joint 1 can be effectively captured. Local discharge usually occurs in the insulating part of the cable joint 1 or the area connected to the cable. Therefore, the discharge signal of the cable joint 1 will be the input source for subsequent signal processing. The function of the mutual inductance acquisition component is to capture the current and voltage fluctuations at the cable joint 1 through a high-frequency pulse current transformer 5 or a power frequency phase transformer 6. The signal output end of the component is connected to the signal input end of the local discharge collector 2. Through this connection, the mutual inductance acquisition component transmits the captured signal to the local discharge collector 2, which is responsible for further processing these signals. The main function of the local discharge collector 2 is to detect whether local discharge occurs at the cable joint 1 through the collected current signal, and amplify, filter and convert the signal for subsequent analysis and classification. The signal output end of the partial discharge collector 2 is connected to the first signal input end of the monitoring host 4. This connection relationship ensures the data transmission of the signal from the partial discharge collector 2 to the monitoring host. After the partial discharge collector 2 processes the signal, the signal will be transmitted to the monitoring host 4 through this channel. After receiving the signal, the monitoring host 4 uses the built-in classification and recognition model to analyze and process the signal to determine the type and location of the discharge signal. This is the core step in the monitoring process and directly affects the identification and positioning of the discharge signal. The environmental data collector 3 is connected to the second signal input end of the monitoring host 4 to collect environmental data related to the discharge signal, such as temperature, humidity, electromagnetic field strength, etc. The environmental data collector 3 collects environmental change data through sensors and transmits it to the monitoring host 4 through this connection. The acquisition of environmental data is crucial to compensate for environmental influences in the analysis of discharge signals, because environmental conditions (such as temperature, humidity, electromagnetic interference, etc.) have a significant impact on the propagation and detection of discharge signals. The monitoring host 4 uses these environmental data, combined with partial discharge signals, to further improve the precision and accuracy of the discharge source positioning. Through the connection between the above components, the entire cable joint partial discharge online monitoring device forms a complete workflow: the discharge signal is collected from the cable joint 1, the signal is captured by the mutual inductance acquisition component, and the signal is transmitted to the partial discharge collector 2 for processing, and finally the monitoring host 4 analyzes the signal and outputs the information of the discharge type and location. At the same time, the environmental data collector 3 provides relevant environmental data to assist the monitoring host 4 in performing more accurate analysis. This system can monitor the status of the cable joint 1 in real time, detect and locate partial discharge problems in a timely manner, and ensure the safe and stable operation of the power system.

[0061] Further, such as Figure 7As shown, the mutual inductance collection component includes a high-frequency pulse current transformer 5, the signal collection end of the high-frequency pulse current transformer 5 is coaxially spaced with the cross-interconnected grounding wire, and the signal output end of the high-frequency pulse current transformer 5 is connected to the signal input end of the partial discharge collector 2.

[0062] In this embodiment, the mutual inductance acquisition component is composed of a high-frequency pulse current transformer 5, which is used to capture the local discharge phenomenon of the cable joint 1 by monitoring the current signal at the cable joint 1. The high-frequency pulse current transformer 5 is used to detect the high-frequency current pulse caused by local discharge, and is mainly used to analyze whether discharge occurs at the cable joint 1 and the intensity of the discharge. The signal acquisition end of the high-frequency pulse current transformer 5 is coaxially spaced with the cross-interconnected grounding wire. This design ensures that the high-frequency pulse current transformer 5 can accurately capture the high-frequency current signal on the cross-interconnected grounding wire connected to the cable joint 1. Through the setting of the coaxial spacing, the high-frequency pulse current transformer 5 can effectively avoid external interference and ensure that the signal related to the local discharge at the cable joint 1 is captured. The position design of the signal acquisition end allows the high-frequency pulse current transformer 5 to sensitively respond to the current fluctuations around the cable joint 1, especially the instantaneous pulses generated by the discharge event, to ensure the accurate capture of the high-frequency current signal. The signal output end of the high-frequency pulse current transformer 5 is connected to the signal input end of the local discharge collector 2, and the signal is transmitted from the high-frequency pulse current transformer 5 to the local discharge collector 2 through this channel. In this connection, the high-frequency pulse current transformer 5 outputs the collected current signal to the partial discharge collector 2, which will further process the signal. The task of the partial discharge collector 2 is to amplify, filter and convert the original signal from the high-frequency pulse current transformer 5 into analog-to-digital conversion, thereby generating data suitable for analysis. The partial discharge collector 2 not only amplifies the signal to make it suitable for subsequent analysis, but also removes noise through filtering, retains useful partial discharge signals, and converts these signals into digital format for further processing by the monitoring host 4. Through such data flow, the system can detect the state of the cable joint 1 in real time and determine whether there is a partial discharge phenomenon. The signal collection and processing chain of the high-frequency pulse current transformer 5 ensures the high accuracy and high reliability of the monitoring system, and provides accurate basic data for subsequent discharge type identification and location judgment. In general, the connection relationship of this component ensures that the high-frequency current signal collected from the cable joint 1 can be accurately transmitted to the partial discharge collector 2, and further signal processing is provided for analysis by the monitoring host 4. Through this process, the system can effectively detect and locate the partial discharge phenomenon at the cable joint 1, and provide key data for troubleshooting.

[0063] Further, such as Figure 7As shown, the mutual inductance collection component also includes an industrial frequency phase mutual inductor 6, the signal collection end of the industrial frequency phase mutual inductor 6 is coaxially spaced with the cable, and the signal output end of the industrial frequency phase mutual inductor 6 is connected to the signal input end of the partial discharge collector 2.

[0064] In this embodiment, the mutual inductance acquisition component also includes a power frequency phase mutual inductor 6, which is used to monitor the power frequency signal in the cable system. The power frequency phase mutual inductor 6 can obtain the phase information of the cable during operation by monitoring the power frequency signal of the cable, and assist in detecting the electrical disturbance caused by discharge. The power frequency signal generally matches the operating frequency of the power system, usually 50Hz or 60Hz. The signal acquisition end of the power frequency phase mutual inductor 6 is coaxially spaced with the cable. This design ensures that the power frequency phase mutual inductor 6 can accurately capture the power frequency current signal in the cable. Through the setting of the coaxial spacing, the power frequency phase mutual inductor 6 can accurately sense the current signal of the cable while avoiding the influence of other environmental interference on the measurement results. In this way, the power frequency phase mutual inductor 6 can obtain the signal in the cable, especially the signal characteristics related to the power frequency of the power system, and provide key data for subsequent analysis. The signal output end of the power frequency phase mutual inductor 6 is connected to the signal input end of the partial discharge collector 2. The function of this connection relationship is to transmit the power frequency signal collected by the power frequency phase mutual inductor 6 to the partial discharge collector 2. The partial discharge collector 2 is responsible for further processing these signals, including processing steps such as amplification, filtering and analog-to-digital conversion. The partial discharge collector 2 can extract useful information from the signal and convert it into a format suitable for subsequent analysis. For example, the phase information in the power frequency signal can help identify the operating status of the cable and whether it is affected by partial discharge. The power frequency phase signal combined with the partial discharge signal can provide comprehensive diagnostic information of the cable fault, especially when the power frequency disturbance is detected, it can more accurately determine whether there is a partial discharge phenomenon. Through this signal transmission and processing chain, the system can simultaneously obtain the characteristics of high-frequency and power frequency signals, thereby providing a more comprehensive perspective for the monitoring of partial discharge. This method can not only identify the high-frequency component of the discharge signal, but also combine it with the power frequency signal to further improve the accuracy of the discharge type and location determination, thereby achieving more accurate fault detection and positioning.

[0065] Further, such as Figure 7 As shown, the monitoring host 4 is an FPGA processor.

[0066] In this embodiment, first, the FPGA processor has a high degree of parallel processing capability. FPGA (field programmable gate array) is an integrated circuit that can be programmed according to specific needs, and its architecture can realize parallel processing of multiple tasks. Compared with traditional processors (such as CPU), FPGA can process multiple signal channels at the same time and perform multi-task parallel calculations. In the online monitoring system for partial discharge of cable joints, FPGA can quickly and simultaneously process data streams from multiple sensors during real-time signal processing, such as high-frequency pulse current signals and power frequency phase signals, thereby greatly improving the data processing speed and system responsiveness. Especially in the process of signal acquisition and analysis, the parallel processing capability of FPGA can ensure that the system can process a large amount of data in real time and respond quickly without increasing delay. Secondly, the FPGA processor has flexible hardware acceleration capabilities. The hardware resources of FPGA can be customized and configured according to specific application requirements, allowing users to perform hardware-level acceleration for specific algorithms. For example, in the partial discharge monitoring system for cable joints, FPGA can implement customized signal filtering, discharge type classification and location positioning algorithms, and accelerate these processes directly at the hardware level without the need for software implementation. This hardware acceleration enables FPGAs to complete complex computing tasks more efficiently than traditional processors, especially in application scenarios that require high-speed real-time processing. FPGAs can significantly reduce processing delays and improve the real-time and accuracy of the system. In addition, FPGA processors are highly programmable and flexible. FPGAs allow users to reconfigure or program hardware according to actual needs, allowing the system to flexibly respond to different application requirements. In the cable joint partial discharge monitoring system, FPGAs can dynamically adjust algorithms according to monitoring requirements and changes in discharge signals to adapt to signal processing requirements in different environments. Whether it is complex filtering of signals or fusion of multi-dimensional data within a specific time, FPGAs can be optimized according to actual conditions, allowing the system to provide stable performance in different situations. Finally, FPGA processors have low power consumption and efficient performance. Compared with traditional processors or microprocessors, FPGAs generally have higher computing efficiency and lower power consumption. For monitoring systems that need to run continuously for a long time, FPGAs can reduce the energy consumption of the system while ensuring efficient processing. Reducing power consumption is particularly important when they need to be deployed on-site for long-term monitoring. The low power consumption characteristics of FPGAs make them suitable for power equipment monitoring systems that require long-term stable operation.

[0067] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0068] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for online monitoring of partial discharge of cable joints, characterized in that: The method for online monitoring of partial discharge of a cable joint comprises: If a discharge pulse signal different from the background noise is detected, determining a discharge type determination feature and a discharge position determination feature of the discharge pulse signal; Based on the established classification and recognition model, the discharge type determination features are analyzed to determine the corresponding discharge type; according to the discharge type, the corresponding environmental sensor element combination is triggered to obtain the corresponding environmental data; Determining a corresponding discharge position according to the environmental data and the discharge position determination feature; A corresponding monitoring result is generated based on the association between the discharge position and the discharge type, and the monitoring result is pushed to a corresponding online terminal.

2. A cable joint partial discharge online monitoring method according to claim 1, characterized in that: If a discharge pulse signal different from background noise is detected, the step of determining a discharge type determination feature and a discharge position determination feature of the discharge pulse signal includes: If a discharge pulse signal different from the background noise is detected, the corresponding sampling period is determined; Based on the channel waveform diagram, determining the time characteristics, frequency characteristics, amplitude characteristics and phase characteristics within the sampling period, and determining the time characteristics, frequency characteristics, amplitude characteristics and phase characteristics as discharge type determination characteristics; Based on the double-terminal positioning measurement method, the corresponding discharge position determination feature is determined by calculating the time difference between the discharge pulse signal reaching different sensors within the sampling period.

3. A cable joint partial discharge online monitoring method according to claim 2, characterized in that: The step of analyzing the discharge type determination feature based on the established classification and recognition model to determine the corresponding discharge type includes: Matching the type weight coefficient corresponding to each of the discharge type determination features, and generating a corresponding type feature vector according to the product of the type weight coefficient and the discharge type determination feature; Adding each of the type of feature vectors to generate a corresponding comprehensive feature vector; Substitute the comprehensive feature vector into the established classification and recognition model to output the corresponding discharge type.

4. A cable joint partial discharge online monitoring method according to claim 2, characterized in that: The step of triggering a corresponding combination of environmental sensor elements to acquire corresponding environmental data according to the discharge type includes: Determining corresponding associated dimensions according to the discharge type, and determining a triggering order of each associated dimension; The corresponding sensor elements are matched according to the associated dimensions, and the various sensor elements are integrated into a corresponding environmental sensor element combination; based on the trigger order, the corresponding sampling interval is determined within the sampling period, and according to the trigger order and the sampling interval, the corresponding environmental sensor element combination is triggered to obtain the corresponding environmental data.

5. The method for online monitoring of partial discharge of cable joints according to claim 1, characterized in that: The step of determining the corresponding discharge position according to the environmental data and the discharge position determination feature includes: Preprocessing the environmental data and the discharge position determination features; Based on regression analysis, determine the compensation factor i corresponding to the environmental data i, wherein the environmental data i is the i-th environmental data, and the compensation factor i is the influence coefficient of the i-th environmental data on the discharge position; Determining an initial discharge position S1 according to the discharge position determination feature; The environmental data i, compensation factor i and initial discharge position S1 are substituted into the established compensation formula to generate a compensated discharge position S2. The established compensation formula is: S2=S1+∑(compensation factor i×environmental data i).

6. The method for online monitoring of partial discharge of cable joints according to claim 1, characterized in that: In the step of generating a corresponding monitoring result based on the association between the discharge position and the discharge type, the monitoring result at least includes an alarm instruction and a visualization model, and the step further includes: According to the discharge position, determining a corresponding associated impact area; Determining, according to the discharge type, an association result of the association impact area; Determine whether the correlation result exceeds a preset result threshold, if not, continue to determine, if exceeded, generate a corresponding alarm instruction; A corresponding visualization model is established according to the association impact area and the association result.

7. An online monitoring device for partial discharge of cable joints, characterized in that: A cable joint partial discharge online monitoring method as claimed in any one of claims 1 to 6 is used, wherein the cable joint partial discharge online monitoring device comprises a mutual inductance acquisition component, a cable joint (1), a partial discharge collector (2), an environmental data collector (3) and a monitoring host (4), wherein the cable joint (1) is sleeved on the cable and connected to the cross-connected grounding wire, the signal output end of the mutual inductance acquisition component is connected to the signal input end of the partial discharge collector (2), the signal output end of the partial discharge collector (2) is connected to the first signal input end of the monitoring host (4), and the signal output end of the environmental data collector (3) is connected to the second signal input end of the monitoring host (4).

8. The on-line monitoring device for partial discharge of cable joints according to claim 7 is characterized in that: The mutual inductance collection component comprises a high-frequency pulse current transformer (5), a signal collection end of the high-frequency pulse current transformer (5) is coaxially spaced from the cross-connected grounding wire, and a signal output end of the high-frequency pulse current transformer (5) is connected to a signal input end of the partial discharge collector (2).

9. The on-line monitoring device for partial discharge of cable joints according to claim 7, characterized in that: The mutual inductance collection component also includes an industrial frequency phase mutual inductor (6), the signal collection end of the industrial frequency phase mutual inductor (6) is coaxially spaced from the cable, and the signal output end of the industrial frequency phase mutual inductor (6) is connected to the signal input end of the partial discharge collector (2).

10. The on-line monitoring device for partial discharge of cable joints according to claim 7, characterized in that: The monitoring host (4) is a FPGA processor.

Images