A method and device for online monitoring of partial discharge of cable joints

Abstract

The present application relates to a method and device for online monitoring of partial discharge of cable joints. The method comprises determining the discharge type determination feature and the discharge position determination feature of the discharge pulse signal; analyzing the discharge type determination feature to determine the corresponding discharge type; triggering the corresponding environmental sensor element combination to obtain the corresponding environmental data; determining the corresponding discharge position according to the environmental data and the discharge position determination feature; analyzing the extracted discharge type determination feature to accurately classify the type of discharge, and the environmental data is used as a compensation factor to help further optimize the discharge position determination feature. In combination with these environmental data and discharge position determination features, the system adopts a multimodal data fusion algorithm to accurately calculate the specific location of the discharge source. Through the multi-dimensional comprehensive analysis of signals, environmental data and locations, the accuracy and reliability of discharge monitoring are significantly improved.

Description

Technical Field

[0001] The present 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] Currently, 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 primarily use discharge pulse signals detected by sensors, using simple threshold determination and statistical analysis methods to classify discharges and subsequently assess the condition of cable joints.

[0003] Existing methods for monitoring partial discharges at cable joints have limitations. Traditional methods rely on a single signal detection and classification model, making it difficult to accurately distinguish different types of partial discharge signals. This can easily lead to false alarms or missed alarms, especially in the presence of complex noise or environmental interference. Furthermore, existing monitoring methods often fail to adjust monitoring strategies based on different discharge types, limiting monitoring accuracy and response speed. Furthermore, precisely locating the discharge location remains a technical challenge, and existing methods often struggle 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:

[0006] If a discharge pulse signal that is different from background noise is detected, determining a discharge type determination feature and a discharge position determination feature of the discharge pulse signal;

[0007] 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;

[0008] determining a corresponding discharge position according to the environmental data and the discharge position determination feature;

[0009] 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 the corresponding online terminal.

[0010] By adopting the above-mentioned technical solution, the present application automatically identifies valid discharge signals that are different from background noise by comparing the detected discharge pulse signal with the background noise. On this basis, it extracts discharge type determination features, such as time, frequency, amplitude, and phase characteristics, as well as discharge location determination features, such as signal arrival time difference. Then, based on the established classification and recognition model, the extracted discharge type determination features are analyzed to accurately classify the discharge type, such as tip discharge, surface discharge, etc. Next, based on 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. This environmental data is used as a compensation factor to help further optimize the discharge location determination features. Combining these environmental data and discharge location 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 via the communication network, allowing operators to 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.

[0011] Preferably, if a discharge pulse signal that is 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:

[0012] If a discharge pulse signal that is different from background noise is detected, the corresponding sampling period is determined;

[0013] 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;

[0014] 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.

[0015] 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 the 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 integrate multi-dimensional information for feature extraction, 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.

[0016] Preferably, the step of analyzing the discharge type determination features based on the established classification and recognition model to determine the corresponding discharge type includes:

[0017] 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;

[0018] Adding the characteristic vectors of each type to generate a corresponding comprehensive characteristic vector;

[0019] Substitute the comprehensive feature vector into the established classification and recognition model to output the corresponding discharge type.

[0020] 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 classification accuracy; by adding the type feature vectors to generate a comprehensive feature vector, multiple features can be fused, thereby improving the recognition ability of the model and reducing 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.

[0021] Preferably, the step of triggering a corresponding combination of environmental sensor elements to obtain corresponding environmental data according to the discharge type includes:

[0022] Determining corresponding associated dimensions according to the discharge type, and determining a triggering order for each associated dimension;

[0023] 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 triggering order, the corresponding sampling interval is determined within the sampling period, and according to the triggering order and the sampling interval, the corresponding environmental sensor element combination is triggered to obtain the corresponding environmental data.

[0024] 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 environmental data collection is carried out at the most appropriate time, thereby improving the response speed and accuracy of the system.

[0025] Preferably, the step of determining the corresponding discharge position according to the environmental data and the discharge position determination feature includes:

[0026] Preprocessing the environmental data and the discharge position determination features;

[0027] Based on the regression analysis, determining 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;

[0028] Determining an initial discharge position S1 according to the discharge position determination feature;

[0029] 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).

[0030] By adopting the above technical solution, by preprocessing the environmental data and discharge position determination features, noise can be eliminated and the data scale can be unified, thereby ensuring the accuracy of subsequent analysis; by determining the compensation factor corresponding to the environmental data based on regression analysis, the impact of environmental factors on the discharge position can be quantified, 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.

[0031] 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:

[0032] Determining a corresponding associated impact area according to the discharge position;

[0033] determining, according to the discharge type, a correlation result of the correlation impact area;

[0034] Determine whether the correlation result exceeds a preset result threshold, if not, continue to determine, if exceeded, generate a corresponding alarm instruction;

[0035] A corresponding visualization model is established according to the association impact area and the association result.

[0036] 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 association result 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 association result exceeds the preset threshold, an alarm instruction is generated if it exceeds, and an alarm can be issued in time in the event of abnormal discharge to ensure the safety of the system; by establishing a visualization model based on the associated impact area and the association result, the monitoring results can be intuitively displayed, helping operators to make decisions quickly.

[0037] A device for online monitoring partial discharge of a cable joint uses a method for online monitoring partial discharge of a cable joint. The 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.

[0038] By adopting the above technical solution, 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 nature 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.

[0039] Preferably, the mutual inductance acquisition component includes a high-frequency pulse current transformer, the signal acquisition 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.

[0040] By adopting the above technical solution and applying a 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 partial discharge collector, the signal data can be efficiently transmitted, ensuring the stability and reliability of data acquisition.

[0041] Preferably, the mutual inductance acquisition component further includes a power frequency phase mutual inductor, the signal acquisition 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.

[0042] 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 partial discharge collector, the precision and accuracy of signal acquisition can be further improved, ensuring the accurate determination of the discharge type and location.

[0043] Preferably, the monitoring host is an FPGA processor.

[0044] 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, realize efficient processing of large-scale data, and ensure the real-time and stability of the system.

[0045] In summary, this application includes at least one of the following beneficial technical effects:

[0046] This application automatically identifies effective discharge signals that are different from background noise by comparing the detected discharge pulse signal with the background noise, and on this basis extracts discharge type determination features, such as time, frequency, amplitude and phase features, as well as discharge position determination features, such as signal arrival time difference. 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. These environmental data are used as compensation factors to help further optimize the discharge position determination features. Combining 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

[0047] Figure 1 This is a flow chart of a method for online monitoring of partial discharge in cable joints in one embodiment of the present application.

[0048] 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;

[0049] Figure 3 This is a flowchart for implementing step S20 in a method for online monitoring of partial discharge in cable joints according to one embodiment of the present application;

[0050] Figure 4 This is a flowchart for implementing step S30 in a method for online monitoring of partial discharge in cable joints in one embodiment of the present application;

[0051] 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;

[0052] Figure 6 This 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;

[0053] Figure 7 It is a schematic diagram of the specific structure of an online monitoring device for partial discharge of cable joints in one embodiment of the present application. DETAILED DESCRIPTION

[0054] The present application is further described in detail below with reference to the accompanying drawings.

[0055] In one embodiment, if Figure 1 As shown, the present application discloses an online monitoring method for partial discharge of cable joints, which specifically includes the following steps:

[0056] S10, if a discharge pulse signal different from background noise is detected, determining a discharge type determination feature and a discharge position determination feature of the discharge pulse signal;

[0057] In this embodiment, if a discharge pulse signal distinct from background noise is detected, the system uses precise signal processing methods to filter and eliminate noise from the input signal, ensuring that only valid discharge signals are retained. During this process, the system first uses a bandpass filter to remove low- and high-frequency noise from the signal, ensuring that the signal is concentrated within the frequency band of partial discharge. The system then denoises the signal, utilizing noise estimation and adaptive filtering techniques to separate the true discharge signal from the ambient noise. This series of processes effectively improves the signal-to-noise ratio, ensuring that subsequent analysis relies solely on valid discharge signals, thereby enhancing the accuracy and stability of discharge signal identification.

[0058] S20, analyzing the discharge type determination features based on the established classification and recognition model to determine the corresponding discharge type;

[0059] In this embodiment, the discharge type determination features are analyzed based on the established classification and 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.

[0060] S30, triggering a corresponding combination of environmental sensing elements to obtain corresponding environmental data according to the discharge type;

[0061] In this embodiment, based on the type of discharge, the system triggers the corresponding combination of environmental sensor elements to acquire environmental data. Depending on the different types of discharge, the system automatically selects appropriate environmental sensors for data acquisition. For example, when a tip discharge is detected, the system will prioritize activating the electromagnetic field strength sensor, because 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 changes in temperature and humidity. After triggering the corresponding combination of environmental sensor elements, the system can collect accurate environmental data, such as electromagnetic field strength, temperature, humidity, etc., in real time, providing rich environmental information to assist in subsequent analysis. This process enables the monitoring system to dynamically adjust the data acquisition strategy for different discharge types, thereby improving the accuracy and adaptability of monitoring.

[0062] S40, determining a corresponding discharge position according to the environmental data and the discharge position determination feature;

[0063] In this embodiment, the system is able to determine the exact discharge location by combining environmental data and discharge position determination features. By fusing environmental data (such as temperature and humidity, electromagnetic fields, etc.) with characteristics such as the time difference and signal strength of the discharge location, the system can accurately locate the location 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, ensuring more accurate discharge position positioning.

[0064] 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.

[0065] In this embodiment, 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 associated information of the potential fault area. The monitoring results can be displayed through a graphical interface, such as using a heat 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.

[0066] 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:

[0067] S101, if a discharge pulse signal different from background noise is detected, determining a corresponding sampling period;

[0068] 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 real-time signal strength and frequency analysis, the system can identify meaningful discharge pulses and set an appropriate sampling period 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 the signal strength is detected to reach a predetermined threshold, the system will automatically set an adaptive sampling period that is long enough to cover the complete discharge signal to avoid the collection effect being affected by environmental noise or interference signals.

[0069] S102: determining the time characteristics, frequency characteristics, amplitude characteristics, and phase characteristics within a sampling period based on the channel waveform, and determining the time characteristics, frequency characteristics, amplitude characteristics, and phase characteristics as discharge type determination characteristics;

[0070] In this embodiment, based on the channel waveform, the time, frequency, amplitude, and phase characteristics within the sampling period are determined and identified as discharge type determination features. In this step, the system analyzes the waveform signal collected by the sensor and extracts four key features: time, frequency, amplitude, and phase. Specifically, time features refer to the signal's duration, pulse interval, and start and end times, which help determine the periodicity and stability of the discharge. Frequency features, obtained by Fourier transforming the signal, reveal its spectral distribution and reflect the frequency range of the discharge. Tip discharges typically appear as high-frequency signals, while surface discharges may appear as lower-frequency signals. Amplitude features measure signal strength and indicate the magnitude of the discharge. Different types of discharge signals can vary significantly in strength, and signals with larger amplitudes may indicate serious faults. Phase features describe the phase variation of the signal within a time period. Tip discharges typically occur at specific phases of the power grid cycle and have significant phase characteristics. These features are extracted through mathematical models and used in subsequent discharge type determination, improving the accuracy of discharge type identification and helping to distinguish between different discharge patterns.

[0071] 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.

[0072] In this embodiment, based on a two-terminal positioning measurement method, the corresponding discharge location determination feature is determined by calculating the time difference between the discharge pulse signal's arrival at different sensors within a sampling period. By installing sensors at both ends of the cable or at multiple locations, the system can capture the time difference in the discharge signal's propagation. Specifically, the system uses a two-terminal positioning measurement method to calculate the location of the discharge source by measuring the propagation time difference between the discharge signal from the discharge source to the different sensors. Since the propagation speed of electromagnetic signals in different media is known, by accurately measuring the time difference between the signal's arrival at each sensor, the system can accurately calculate the coordinates or area of ​​the discharge source. The time difference in the arrival of the discharge signal is proportional to the distance. Utilizing this principle, the system can use an algorithm to calculate the distance between the discharge source and the sensor, ultimately determining the exact location of the discharge source. Furthermore, the system can compensate for this time difference using environmental data (such as temperature, humidity, and electromagnetic interference) to further improve positioning accuracy.

[0073] 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:

[0074] 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;

[0075] In this embodiment, the type weight coefficient corresponding to each discharge type determination feature is matched, and the corresponding type feature vector is generated based on 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 signal's time characteristics (pulse width, pulse interval), frequency characteristics (signal spectrum, frequency distribution), amplitude characteristics (signal strength, amplitude change), and phase characteristics (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. To improve classification accuracy, the system assigns a weight coefficient to each feature. The values ​​of these weight coefficients reflect the contribution of different features to the discharge type determination. The weight coefficients are derived 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 tip discharge typically manifests as a high-frequency signal, while for surface discharge, the time and amplitude features 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.

[0076] S202, adding the feature vectors of each type to generate a corresponding comprehensive feature vector;

[0077] In this embodiment, each type of feature vector is added together 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, time, frequency, amplitude and phase features may describe the type of discharge from different perspectives, 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 judgment accuracy of the model.

[0078] S203: Substitute the comprehensive feature vector into the established classification and recognition model, and output the corresponding discharge type.

[0079] In this embodiment, the comprehensive feature vector is substituted into the established classification and recognition model to output the corresponding discharge type. 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), decision tree, random forest, 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 continuously adjusts internal parameters so that the model can accurately classify new discharge signals in actual applications. In the process of outputting the discharge type, the model weights each feature according to the feature weights of different discharge types to obtain the most appropriate classification result, thereby improving the accuracy of discharge type recognition and the reliability of the system.

[0080] 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.

[0081] 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 corresponding environmental data according to the discharge type, the following steps are included:

[0082] S301, determining corresponding associated dimensions according to the discharge type, and determining a triggering order for each associated dimension;

[0083] In this embodiment, based on 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 based on different types of discharge signals. 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 based on 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 the triggering order of these associated dimensions based on their characteristics, ensuring that the collected environmental data can be synchronized with the characteristics of the discharge signal, further improving the timeliness of monitoring.

[0084] S302, matching corresponding sensor elements according to the associated dimensions, and integrating the sensor elements into a corresponding environmental sensor element combination;

[0085] In this embodiment, corresponding sensor elements are matched according to associated dimensions, and the individual sensor elements are integrated into corresponding environmental sensor element combinations. Different discharge types require different environmental data to compensate for and support judgment. Therefore, the system selects the most appropriate sensor element for each dimension based on the identified associated dimensions. For example, if the associated dimensions are temperature and humidity, the system may select a temperature and humidity sensor; if the associated dimensions are electromagnetic field strength, the system may select an electromagnetic field sensor. After selecting the sensors, 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 that can simultaneously obtain environmental data from multiple dimensions, providing sufficient information for accurate judgment of the discharge type. In this way, the system can dynamically select and activate sensors based on the actual discharge type requirements, avoiding unnecessary sensor waste and improving system resource utilization and data collection efficiency.

[0086] S303 : Based on the triggering sequence, 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 sequence and the sampling interval.

[0087] In this embodiment, based on the trigger sequence, the corresponding sampling interval is determined within the sampling period. According to the trigger sequence and sampling interval, the corresponding environmental sensor element combination is triggered to obtain the corresponding environmental data. The system determines the activation time of different sensor element combinations based on the aforementioned trigger sequence to ensure that different environmental data are collected at the optimal time. The design of the trigger sequence 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 between sensor activations) to ensure the accuracy and timeliness of data collection. The sampling interval is dynamically adjusted based on the duration of the discharge signal, the speed of environmental changes, and monitoring requirements to avoid excessive or missed environmental data collection. In this way, the system not only improves the efficiency of data collection, but also ensures that the collected data accurately reflects environmental changes, further optimizing the judgment and analysis of discharge types.

[0088] Specifically, suppose the system detects a tip discharge signal. Based on the characteristics of the tip discharge, the system identifies associated environmental dimensions, such as electromagnetic field strength and temperature and humidity. These two dimensions are crucial for monitoring tip discharges, as they are often accompanied by strong electromagnetic field fluctuations, while changes in ambient temperature and humidity can affect the discharge's intensity and propagation characteristics. The system selects these two dimensions as relevant features based on the type of discharge and sets their triggering order based on experience or pre-set rules. First, the system determines that electromagnetic field strength is the most critical associated dimension, so the electromagnetic field sensor is triggered first. To ensure timely acquisition of electromagnetic field data during the discharge process, the system sets a short time interval for rapid electromagnetic field data collection. The temperature and humidity sensors are triggered more slowly, with a slightly longer sampling interval. This sampling strategy ensures that electromagnetic field data closely tracks changes in the discharge event, while temperature and humidity data are collected more smoothly, avoiding excessive interference with data collection. The system then activates the electromagnetic field strength sensor based on this triggering order and collects data. Assuming that during the discharge process, the fluctuation of electromagnetic field intensity reaches its maximum near the signal peak, the system captures this changing electromagnetic field data and combines it 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 this environmental data, the system ultimately confirms the type of tip discharge and accurately calculates the location of the discharge source. Through 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, real-time changes in electromagnetic field intensity are crucial for determining tip discharge, while changes in temperature and humidity provide supplementary information on the long-term impact of discharge. This mechanism of dynamically adjusting the triggering of environmental sensors effectively improves the system's adaptability to different discharge types, thereby providing more accurate and reliable partial discharge monitoring in practical applications.

[0089] 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, the following steps are included:

[0090] S401, pre-processing environmental data and discharge position determination features;

[0091] In this embodiment, the environmental data and discharge position determination features are preprocessed. The preprocessing step is the process of cleaning and optimizing the raw 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 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 slight errors due to environmental fluctuations, or data drift caused by sensor aging; 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 calculation and precise positioning of the discharge position, thereby improving the accuracy and stability of the entire system.

[0092] 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;

[0093] In this embodiment, a compensation factor i corresponding to environmental data i is determined based on regression analysis, where environmental data i is the i-th environmental data and compensation factor i is the coefficient of influence of the i-th environmental data on the discharge location. The core of this step is to quantify the impact of different environmental factors on the discharge location using a regression analysis model. The regression analysis model is built based on historical or experimental data and incorporates the relationships between multiple environmental variables and the discharge location. Specifically, the system performs regression analysis on existing environmental data and actual discharge location data to identify which environmental data (such as temperature, humidity, and electromagnetic field strength) have a strong correlation with the discharge location offset. For example, the insulation material of a cable may expand or contract at different temperatures, causing changes in the speed of signal propagation, thereby affecting the accuracy of the discharge location. Through regression analysis, the system assigns a compensation factor to each type of environmental data. This compensation factor reflects the degree of influence of the environmental data on the discharge location. For example, temperature may have a higher impact on the discharge location, while humidity may have a lower impact. This compensation factor can dynamically adjust and compensate for errors caused by environmental changes, ensuring more accurate discharge location calculations.

[0094] S403, determining an initial discharge position S1 according to the discharge position determination feature;

[0095] In this embodiment, the initial discharge position S1 is determined based on the discharge position determination feature. The discharge position determination feature is usually a preliminary position calculated by measuring the time difference between the arrival of the signal at 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 the signal propagation. The time difference between the arrival of the signal at the sensor reflects the relative distance between the discharge source and each sensor. These time differences will be used to infer the preliminary value of the discharge position through triangulation or multi-point positioning algorithms. Due to factors such as environmental interference and sensor errors, 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 between the discharge source and the two sensors based on the known signal propagation speed, thereby deriving a preliminary discharge position S1. This is a preliminary estimated position, and it still needs to be compensated by environmental data to improve accuracy.

[0096] S404 , substituting the environmental data i, the compensation factor i and the initial discharge position S1 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).

[0097] 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 using the compensation formula, taking into account the impact of environmental factors on position determination. The compensation formula is a mathematical model, usually derived 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, assuming 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 together with 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 impact 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.

[0098] 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:

[0099] S501, determining a corresponding associated impact area according to a discharge position;

[0100] In this embodiment, the associated impact area is determined based on the discharge location. This step precisely calculates the location of the discharge source and determines the potentially affected area based on the equipment layout and power system structure. Specifically, the system uses the discharge location data, combined with the cable geometry, equipment connection structure, and electrical equipment layout, to infer the potential impact area of ​​the discharge. For example, if the discharge occurs near a cable joint, the system may define a circular area centered on the joint as the impact area. If other cables or electrical equipment surrounding the equipment may also be affected by the discharge, the system includes these areas. To achieve this, the system first calculates the distance from the discharge source and the potential impact area for each device or area based on the known relative positions of the discharge source and other equipment. The size and extent of the impact area vary depending on the intensity, propagation method, and spatial distribution of the discharge type; stronger discharge types have larger impact areas. This method allows the system to accurately identify areas affected by the discharge, providing a concrete basis for subsequent alarm generation and maintenance decisions.

[0101] S502: Determine the correlation result of the correlation impact area according to the discharge type;

[0102] In this embodiment, the association results of the associated impact areas are determined based on the discharge type. This step calculates 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 based on 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 tip discharge exceeds the safety standard, the system will calculate the potential impact of electromagnetic interference on the equipment, thereby determining the association results for the area. Based on these calculations, the system can generate an impact result for each associated area. These results reflect the impact of the discharge type on each area, helping operators to assess the potential risk of failure.

[0103] S503: Determine whether the correlation result exceeds a preset result threshold. If not, continue to determine. If exceeded, generate a corresponding alarm instruction.

[0104] In this embodiment, the correlation result is determined to see if it exceeds a preset threshold. If not, the determination continues. If it does, a corresponding alarm is generated. This step aims to determine whether urgent action is necessary based on the discharge type and the calculated impact result. The system first assesses the threat posed by the current discharge to the equipment by comparing the correlation result with a pre-set safety threshold. If the correlation result does not exceed the preset threshold, the system continues to monitor the cable and equipment status, maintaining real-time monitoring of environmental data. If the correlation result exceeds the threshold, the system deems the discharge to have exceeded a safe range, posing a risk of potential equipment damage or failure, and triggers an alarm. This alarm can be a flashing warning icon on the system interface, an audible alarm, or sent to the operator via text message or email. This alarm mechanism helps operators stay informed of equipment status and take measures to prevent further damage or escalation of the fault. For example, if the discharge location and electromagnetic interference calculations indicate that the equipment may be seriously affected, the system will automatically trigger an alarm, prompting the operator to check the health of the cable connector or surrounding equipment.

[0105] S504: Establish a corresponding visualization model according to the associated impact area and the associated result.

[0106] In this embodiment, a corresponding visualization model is established based on the associated impact areas and correlation results. The core of this step is to graphically display the calculation results and associated data, allowing operators to intuitively understand the impact scope 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 formats may include heat maps, two-dimensional, or three-dimensional models, illustrating the location of the discharge source and the impact of the discharge on equipment and areas. For example, the system can use a heat map to display the electromagnetic interference intensity near a cable joint, with areas of higher intensity highlighted in red and areas of lower intensity highlighted in green. Alternatively, a three-dimensional map can be used to depict the spatial location of cable joints and other equipment, with markers of varying colors or intensities indicating the location of the discharge source and the impact area. This visualization allows operators to quickly identify fault areas requiring priority, improving fault response efficiency and decision-making. Furthermore, the system can generate detailed reports or data charts for each impact area to assist management with subsequent maintenance and analysis.

[0107] 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.

[0108] In this embodiment, the cable joint 1 is a key component of the discharge monitoring system and is installed directly at the cable joint. The primary function of the cable joint 1 is to provide the source of discharge signals for the entire monitoring system. The cable joint 1 is connected to the cross-connected ground wire to ensure stable electrical connections and effective capture of partial discharge signals at the cable joint 1. Partial discharge typically occurs in the insulated portion of the cable joint 1 or in the area where it connects to the cable. Therefore, the discharge signal from the cable joint 1 serves as the input source for subsequent signal processing. The mutual inductance acquisition component captures current and voltage fluctuations at the cable joint 1 using a high-frequency pulse current transformer 5 or a power-frequency phase transformer 6. The signal output of this component is connected to the signal input of the partial discharge collector 2. Through this connection, the mutual inductance acquisition component transmits the captured signals to the partial discharge collector 2, which is responsible for further processing. The primary function of the partial discharge collector 2 is to detect partial discharge at the cable joint 1 using the collected current signals and to amplify, filter, and convert the signals for subsequent analysis and classification. The signal output of the partial discharge collector 2 is connected to the first signal input of the monitoring host 4. This connection ensures data transmission from the partial discharge collector 2 to the monitoring host. After processing the signal, the partial discharge collector 2 transmits it to the monitoring host 4 through this channel. Upon receiving the signal, the monitoring host 4 uses its built-in classification and recognition model to analyze and process the signal to determine the type and location of the discharge signal. This is a core step in the monitoring process and directly impacts the identification and location of the discharge signal. The environmental data collector 3 is connected to the second signal input of the monitoring host 4 to collect environmental data related to the discharge signal, such as temperature, humidity, and electromagnetic field strength. The environmental data collector 3 collects environmental change data using sensors and transmits it to the monitoring host 4 through this connection. Acquiring environmental data is crucial for compensating for environmental influences in discharge signal analysis, as environmental conditions (such as temperature, humidity, and electromagnetic interference) significantly affect the propagation and detection of discharge signals. The monitoring host 4 uses this environmental data, combined with the partial discharge signal, to further improve the precision and accuracy of locating the discharge source. The connections between these components form a complete workflow for the online cable joint partial discharge monitoring device: discharge signals are collected from the cable joint 1, captured by the mutual inductance acquisition component, and transmitted to the partial discharge collector 2 for processing. Finally, the monitoring host 4 analyzes the signals and outputs information on the discharge type and location. Simultaneously, the environmental data collector 3 provides relevant environmental data, assisting the monitoring host 4 in performing more precise analysis. This system monitors the status of the cable joint 1 in real time, enabling the timely detection and location of partial discharge problems, thereby ensuring the safe and stable operation of the power system.

[0109] Further, such as Figure 7As shown, the mutual inductance acquisition component includes a high-frequency pulse current transformer 5, the signal acquisition 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.

[0110] In this embodiment, the mutual inductance acquisition component comprises a high-frequency pulsed current transformer 5, which is used to detect partial discharge (PD) at the cable joint 1 by monitoring the current signal at the cable joint 1. The high-frequency pulsed current transformer 5 is used to detect high-frequency current pulses caused by PD, primarily for analyzing whether a discharge has occurred at the cable joint 1 and the intensity of the discharge. The signal acquisition terminal of the high-frequency pulsed current transformer 5 is coaxially spaced from the cross-connected ground wire. This design ensures that the high-frequency pulsed current transformer 5 can accurately capture the high-frequency current signal on the cross-connected ground wire connected to the cable joint 1. This coaxial spacing effectively prevents external interference and ensures that signals related to PD at the cable joint 1 are captured. The position of the signal acquisition terminal allows the high-frequency pulsed current transformer 5 to sensitively respond to current fluctuations around the cable joint 1, particularly transient pulses generated by a discharge event, ensuring accurate capture of the high-frequency current signal. The signal output terminal of the high-frequency pulsed current transformer 5 is connected to the signal input terminal of the PD collector 2, through which signals are transmitted from the high-frequency pulsed current transformer 5 to the PD collector 2. In this connection, the high-frequency pulsed current transformer 5 outputs the collected current signal to the partial discharge collector 2, which further processes the signal. The partial discharge collector 2 amplifies, filters, and performs analog-to-digital conversion on the raw signal from the high-frequency pulsed current transformer 5, 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, retaining the useful partial discharge signal. It then converts this signal into a digital format for further processing by the monitoring host 4. Through this data flow, the system can monitor the status of the cable joint 1 in real time and determine whether partial discharge is occurring. The signal collection and processing chain of the high-frequency pulsed current transformer 5 ensures the high accuracy and reliability of the monitoring system, providing accurate basic data for subsequent discharge type identification and location determination. Overall, this component connection ensures that the high-frequency current signal collected from the cable joint 1 is accurately transmitted to the partial discharge collector 2 and, after further signal processing, is analyzed by the monitoring host 4. Through this process, the system can effectively detect and locate partial discharge at the cable joint 1, providing critical data for troubleshooting.

[0111] Further, such as Figure 7As shown, the mutual inductance acquisition component also includes a power frequency phase mutual inductor 6, the signal acquisition end of the power frequency phase mutual inductor 6 is coaxially spaced from the cable, and 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.

[0112] In this embodiment, the mutual inductance acquisition component also includes a power frequency phase transformer 6, which monitors the power frequency signal in the cable system. By monitoring the power frequency signal in the cable, the power frequency phase transformer 6 can obtain phase information of the cable during operation and assist in detecting electrical disturbances caused by discharges. The power frequency signal generally matches the operating frequency of the power system, typically 50 Hz or 60 Hz. The signal acquisition terminal of the power frequency phase transformer 6 is coaxially spaced from the cable. This design ensures that the power frequency phase transformer 6 can accurately capture the power frequency current signal within the cable. This coaxial spacing allows the power frequency phase transformer 6 to accurately sense the cable current signal while preventing other environmental interference from affecting the measurement results. This allows the power frequency phase transformer 6 to acquire signals in the cable, particularly signal characteristics related to the power system power frequency, providing critical data for subsequent analysis. The signal output terminal of the power frequency phase transformer 6 is connected to the signal input terminal of the partial discharge collector 2. This connection transmits the power frequency signal collected by the power frequency phase transformer 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 combination of the power frequency phase signal and 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 partial discharge monitoring. 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 location.

[0113] Further, such as Figure 7 As shown, the monitoring host 4 is an FPGA processor.

[0114] In this embodiment, first, the FPGA processor possesses a high degree of parallel processing capability. An FPGA (field programmable gate array) is an integrated circuit that can be programmed according to specific requirements, and its architecture enables parallel processing of multiple tasks. Compared to traditional processors (such as CPUs), FPGAs can simultaneously process multiple signal channels and perform multi-task parallel computing. In an online cable joint partial discharge monitoring system, FPGAs can rapidly and simultaneously process data streams from multiple sensors, such as high-frequency pulse current signals and power-frequency phase signals, during real-time signal processing, significantly improving data processing speed and system responsiveness. Particularly during signal acquisition and analysis, the FPGA's parallel processing capabilities ensure that the system can process large amounts of data in real time and respond quickly without increasing latency. Second, the FPGA processor offers flexible hardware acceleration capabilities. FPGA hardware resources can be customized and configured according to specific application requirements, allowing users to implement hardware-level acceleration for specific algorithms. For example, in a cable joint partial discharge monitoring system, FPGAs can implement customized signal filtering, discharge type classification, and location location algorithms, accelerating these processes directly at the hardware level without requiring software implementation. This hardware acceleration enables FPGAs to complete complex computing tasks more efficiently than traditional processors. Especially in applications requiring high-speed, real-time processing, FPGAs can significantly reduce processing latency, improving the system's real-time performance and accuracy. Furthermore, FPGA processors are highly programmable and flexible. FPGAs allow users to reconfigure or reprogram the hardware based on actual needs, enabling the system to flexibly adapt to diverse application requirements. In cable joint partial discharge monitoring systems, FPGAs can dynamically adjust algorithms based on monitoring requirements and discharge signal changes, adapting to signal processing needs in diverse environments. Whether performing complex signal filtering or fusing multidimensional data within a specific timeframe, FPGAs can optimize based on actual conditions, ensuring stable performance in diverse scenarios. Finally, FPGA processors offer low power consumption and high performance. Compared to traditional processors or microprocessors, FPGAs typically offer higher computational efficiency and lower power consumption. For monitoring systems that require continuous operation over extended periods, FPGAs can reduce energy consumption while ensuring efficient processing. This is particularly important when deployed in the field for long-term monitoring. The low power consumption of FPGAs makes them suitable for power equipment monitoring systems requiring long-term, stable operation.

[0115] Those skilled in the art will clearly understand that for the sake of convenience and brevity 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.

[0116] The above-described embodiments 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, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection 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 that is different from 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, triggering a corresponding combination of environmental sensor elements to obtain corresponding environmental data; determining a corresponding discharge position according to the environmental data and the discharge position determination feature; 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; If a discharge pulse signal that is 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 that is different from 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; The step of analyzing the discharge type determination features 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 the characteristic vectors of each type to generate a corresponding comprehensive characteristic vector; Substitute the comprehensive feature vector into the established classification and recognition model to output the corresponding discharge type.

2. The method for online monitoring of partial discharge of cable joints according to claim 1, characterized in that: The step of triggering a corresponding combination of environmental sensor elements to obtain corresponding environmental data according to the discharge type includes: Determining corresponding associated dimensions according to the discharge type, and determining a triggering order for each associated dimension; Matching corresponding sensor elements according to the associated dimensions, and integrating the sensor elements into corresponding environmental sensor element combinations; Based on the triggering order, a corresponding sampling interval is determined within the sampling period, and according to the triggering order and the sampling interval, a corresponding combination of environmental sensing elements is triggered to acquire corresponding environmental data.

3. 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 the regression analysis, determining 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).

4. 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: Determining a corresponding associated impact area according to the discharge position; determining, according to the discharge type, a correlation result of the correlation 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.

5. An online monitoring device for partial discharge of cable joints, characterized in that: A cable joint partial discharge online monitoring method according to any one of claims 1 to 4 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).

6. The cable joint partial discharge online monitoring device according to claim 5, characterized in that: The mutual inductance acquisition component comprises a high-frequency pulse current transformer (5), a signal acquisition end of the high-frequency pulse current transformer (5) is coaxially spaced from a 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).

7. The on-line monitoring device for partial discharge of a cable joint according to claim 5, characterized in that: The mutual inductance acquisition component further comprises an industrial frequency phase mutual inductor (6), a signal acquisition end of the industrial frequency phase mutual inductor (6) is coaxially spaced from the cable, and a signal output end of the industrial frequency phase mutual inductor (6) is connected to a signal input end of the partial discharge collector (2).

8. The cable joint partial discharge online monitoring device according to claim 5, characterized in that: The monitoring host (4) is an FPGA processor.

Images