Monitoring and Early Warning System for High-Voltage Power Equipment Based on Multi-Parameter Fusion Analysis

Abstract

This invention relates to the field of insulation testing technology, specifically a monitoring and early warning system for high-voltage power equipment based on multi-parameter fusion analysis. The system includes a synchronization calibration module, a stripping feature module, a coupling verification module, a temperature field correlation module, a trend judgment module, and a fusion early warning module. It performs timestamp calibration on multi-source heterogeneous data to obtain a synchronous heterogeneous dataset; performs synchronization interference stripping on high-frequency current signals and transient ground voltage signals to obtain phase amplitude spectrum features and pulse group density energy distribution features; and performs cross-validation to obtain valid partial discharge events; performs multi-dimensional correlation analysis on cable joint temperature data and distributed temperature field data to obtain primary fault analysis results; comprehensively judges deviation and evolution rate to obtain degradation trend conclusions; and fuses the primary fault analysis results and degradation trend conclusions with multi-dimensional information to obtain classified early warning information. This invention can improve the efficiency of high-voltage power equipment monitoring.

Description

Technical Field

[0001] This invention relates to the field of insulation testing technology, and in particular to a monitoring and early warning system for high-voltage power equipment based on multi-parameter fusion analysis. Background Technology

[0002] Existing technologies lack an effective time synchronization and calibration mechanism when processing multi-source heterogeneous data from high-voltage power equipment. The timestamps of different types of data are difficult to align accurately, resulting in the inability to form a collaborative relationship between various monitoring data and to fully reflect the spatiotemporal characteristics of the equipment's operating status, thereby affecting the effectiveness of subsequent data processing and analysis.

[0003] Existing technologies fail to adequately remove interference components from high-frequency current signals and transient ground voltage signals, and their extraction of phase amplitude spectrum features and pulse group density energy distribution features is not precise enough. Furthermore, they lack in-depth application of the electrothermal coupling mechanism of partial discharge in insulation, making it impossible to effectively correlate temperature data with partial discharge events. This results in insufficient accuracy in fault identification and makes it difficult to judge equipment degradation trends based on historical health status, thus failing to provide timely and reliable early warning information. Therefore, improving the efficiency of monitoring high-voltage power equipment has become an urgent problem to be solved. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides a monitoring and early warning system for high-voltage power equipment based on multi-parameter fusion analysis, characterized in that the system includes a synchronous calibration module, a feature stripping module, a coupling verification module, a temperature field correlation module, a trend analysis module, and a fusion early warning module, wherein:

[0005] The synchronization calibration module is used to timestamp the multi-source heterogeneous data of the high-voltage power equipment to obtain the synchronous heterogeneous dataset of the high-voltage power equipment.

[0006] The stripping feature module is used to strip the high-frequency current signal and transient ground voltage signal in the synchronous heterogeneous dataset to obtain the phase amplitude spectrum feature and pulse group density energy distribution feature of the power high-voltage equipment.

[0007] The coupling verification module is used to perform spatiotemporal coupling cross-verification of the phase amplitude spectrum features and the pulse group density energy distribution features to obtain the effective partial discharge events of the high-voltage power equipment.

[0008] The temperature field correlation module is used to perform multidimensional correlation analysis on the cable joint temperature data and distributed temperature field data in the effective partial discharge event based on the electrothermal coupling mechanism of insulation partial discharge in high-voltage power equipment, so as to obtain the primary fault analysis results of the high-voltage power equipment.

[0009] The trend analysis module is used to comprehensively analyze the deviation and evolution rate of the synchronous heterogeneous dataset based on the historical health baseline of the high-voltage power equipment, so as to obtain the deterioration trend conclusion of the high-voltage power equipment.

[0010] The fusion early warning module is used to perform multi-dimensional information fusion on the primary fault analysis results and the deterioration trend conclusions to obtain the classification early warning information of the high-voltage power equipment.

[0011] In a preferred embodiment, when the synchronization calibration module performs timestamp calibration on the multi-source heterogeneous data of the high-voltage power equipment to obtain the synchronous heterogeneous dataset of the high-voltage power equipment, it is specifically used for:

[0012] The system synchronously receives distributed temperature field data of cable lines, cable joint temperature data, high-frequency current signals and traveling wave signals, and contact temperature data of switchgear, infrared thermal imaging data and transient ground voltage signals from the high-voltage power equipment to obtain a multi-source heterogeneous dataset of the high-voltage power equipment.

[0013] Time synchronization and alignment are performed on the multi-source heterogeneous dataset to obtain the multi-source signal set of the high-voltage power equipment;

[0014] The multi-source signal set is serialized and encapsulated to obtain the synchronous heterogeneous data of the high-voltage power equipment.

[0015] In a preferred embodiment, when the feature stripping module performs synchronization interference stripping on the high-frequency current signal and transient ground voltage signal in the synchronous heterogeneous dataset to obtain the phase amplitude spectrum features and pulse group density energy distribution features of the high-voltage power equipment, it is specifically used for:

[0016] The high-frequency current data in the synchronous heterogeneous dataset is subjected to pulse waveform discrimination to obtain the high-frequency current pulse signal of the high-voltage power equipment.

[0017] Pulse event detection is performed on the transient ground voltage data in the synchronous heterogeneous dataset to obtain the transient ground voltage pulse event sequence of the power high-voltage equipment;

[0018] The phase position and current amplitude of the discharge pulse in the high-frequency current signal are correlated and matrixed to construct the phase amplitude spectrum characteristics of the high-voltage power equipment.

[0019] Based on the spatiotemporal clustering characteristics and energy fluctuation characteristics of sudden discharges in the high-voltage power equipment, a differentiated characteristic criterion for the high-voltage power equipment is constructed.

[0020] Based on the aforementioned differential characteristic criteria, clustered energy density statistics are performed on the transient ground voltage pulse event sequence to obtain the pulse group density energy distribution characteristics of the high-voltage power equipment.

[0021] In a preferred embodiment, when the feature stripping module performs clustered energy density statistics on the transient ground voltage pulse event sequence based on the differentiated feature criteria to obtain the pulse group density energy distribution characteristics of the high-voltage power equipment, it is specifically used for:

[0022] Based on the spatiotemporal cluster characteristics, the transient ground voltage pulse event sequence is divided into group boundaries to obtain the sudden pulse group of the power high-voltage equipment;

[0023] Time-domain clustering analysis was performed on the burst pulse groups to obtain the interval distribution and duration distribution of the high-voltage power equipment.

[0024] Evolution pattern analysis of the energy intensity of the sudden pulse group yields the pulse energy intensity distribution and evolution trajectory characteristics of the high-voltage power equipment.

[0025] The pulse group density energy distribution characteristics of the high-voltage power equipment are obtained by cross-latitude fusion of the interval distribution, the duration distribution, the pulse energy intensity distribution, and the evolution trajectory characteristics.

[0026] In a preferred embodiment, when the feature stripping module performs evolution pattern analysis on the energy intensity of the burst pulse group to obtain the pulse energy intensity distribution and evolution trajectory characteristics of the high-voltage power equipment, it is specifically used for:

[0027] Based on the amplitude information of the pulse events in the burst pulse group, energy distribution analysis is performed on the pulse events to obtain the pulse energy intensity distribution of the high-voltage power equipment;

[0028] By analyzing the statistical distribution law of the energy release concentration and discrete characteristics of the pulse energy intensity sequence, a description of the intensity distribution of the high-voltage power equipment is obtained.

[0029] Based on the pulse energy intensity sequence, the energy gradient change of the sudden pulse group is identified to obtain a description of the dynamic evolution trend of the high-voltage power equipment.

[0030] Semantic trajectory evolution is performed on the intensity distribution description and the dynamic evolution trend description to obtain the evolution trajectory characteristics of high-voltage power equipment.

[0031] In a preferred embodiment, when the coupling verification module performs spatiotemporal coupling cross-verification of the phase amplitude spectrum features and the pulse group density energy distribution features to obtain the effective partial discharge event of the high-voltage power equipment, it is specifically used for:

[0032] The overlap degree of the active phase intervals of the phase amplitude spectrum features and the high-density pulse group periods of the pulse group density energy distribution features is determined to obtain the time overlap result of the high-voltage power equipment.

[0033] By performing trend analysis on the phase amplitude spectrum features and the pulse group density energy distribution features, the amplitude evolution sequence and intensity evolution sequence of the high-voltage power equipment are obtained.

[0034] Based on the time coincidence results, the trend consistency of the amplitude evolution sequence and the intensity evolution sequence is judged to obtain the trend comparison conclusion of the high-voltage power equipment.

[0035] Based on the physical coupling relationship criteria of the high-voltage power equipment, the event determination is performed on the time coincidence results and the trend comparison conclusion to obtain the effective partial discharge events of the high-voltage power equipment.

[0036] In a preferred embodiment, when the temperature field correlation module performs multidimensional correlation analysis on cable joint temperature data and distributed temperature field data in the effective partial discharge event based on the electrothermal coupling mechanism of insulation partial discharge in high-voltage power equipment to obtain the primary fault analysis results of the high-voltage power equipment, it is specifically used for:

[0037] Information analysis is performed on the occurrence time and signal source attributes of the effective partial discharge events to obtain the target time and associated spatial location of temperature analysis in the high-voltage power equipment;

[0038] Based on the target time, the rate of change of cable joint temperature data in the synchronous heterogeneous dataset is identified to obtain the transient rise rate of the high-voltage power equipment.

[0039] Based on the associated spatial location, temperature field distribution analysis is performed on the distributed temperature field data in the synchronous heterogeneous dataset to obtain the local thermal focusing phenomenon of the high-voltage power equipment.

[0040] Multidimensional spatiotemporal synergy verification was performed on the effective partial discharge event, the transient rise rate, the local thermal focusing phenomenon, and the abnormal disturbance of the traveling wave time difference signal in the synchronous heterogeneous data set to obtain the primary fault analysis results of the high-voltage power equipment.

[0041] In a preferred embodiment, when the trend analysis module performs a comprehensive analysis of the deviation and evolution rate of the synchronous heterogeneous dataset based on the historical health baseline of the high-voltage power equipment to obtain a conclusion on the deterioration trend of the high-voltage power equipment, it is specifically used for:

[0042] Based on the historical health baseline, the synchronous heterogeneous dataset is state-aligned to obtain the comparison data pair sequence of the high-voltage power equipment;

[0043] The difference degree of the comparison data sequence is described to obtain the deviation description sequence of the high-voltage power equipment;

[0044] Based on the continuous time window of the synchronous heterogeneous dataset, the direction and rate of change of the deviation description sequence are tracked and described to obtain the evolution rate description of the high-voltage power equipment;

[0045] An evolution trajectory is constructed from the deviation description sequence and the evolution rate description to obtain the degradation trajectory of the high-voltage power equipment.

[0046] The location information of the degradation trajectory is mapped to a preset stability threshold boundary to obtain the degradation trend conclusion of the high-voltage power equipment.

[0047] In a preferred embodiment, when the fusion early warning module performs multi-dimensional information fusion on the primary fault analysis results and the degradation trend conclusions to obtain the classification early warning information of the high-voltage power equipment, it is specifically used for:

[0048] By jointly determining the results of the initial fault analysis and the conclusion of the deterioration trend, a fault confirmation conclusion is obtained for the high-voltage power equipment.

[0049] Based on the traveling wave time difference signal in the synchronous heterogeneous dataset, the fault section of the cable line with abnormal fault in the high-voltage power equipment is defined to obtain the preliminary fault section range of the high-voltage power equipment.

[0050] Based on the preliminary fault zone range, multi-source spatial information fusion is performed on the spatial center of the local thermal focusing phenomenon in the high-voltage power equipment, the specific location of the abnormal temperature node, and the location of the dominant signal source of the effective partial discharge event to obtain the physical coordinates of the fault point of the high-voltage power equipment.

[0051] Based on the dominant abnormal signal characteristics in the primary fault analysis results, the nature of the abnormal fault is determined to obtain the fault type of the high-voltage power equipment.

[0052] The fault confirmation conclusion, the physical coordinates of the fault point, and the fault type are encapsulated to obtain the classification and early warning information of the high-voltage power equipment.

[0053] In a preferred embodiment, the formula for calculating the physical coordinates of the fault point is as follows:

[0054] ;

[0055] In the formula, The physical coordinates of the fault point are: The coordinates of the spatial center of the localized thermal focusing phenomenon are: These are the specific location coordinates of the abnormal temperature node. The coordinates are the location coordinates of the dominant signal source of the effective partial discharge event. Assign a preset confidence weight to the thermal focusing location. Assign a preset confidence weight to the temperature node location. The preset confidence weights for the location of the discharge signal source are used.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. This invention performs precise timestamp calibration and synchronous integration of multi-source heterogeneous monitoring data from high-voltage power equipment, efficiently removes interference components from high-frequency current signals and transient ground voltage signals, accurately extracts phase amplitude spectrum features and pulse group density energy distribution features, and screens out effective partial discharge events through spatiotemporal coupling cross-validation, significantly improving the integrity and accuracy of power measurement data, and providing a solid and reliable data source support for equipment condition assessment and fault diagnosis.

[0058] 2. This invention relies on the electrothermal coupling mechanism of partial discharge in insulation to deeply correlate effective partial discharge events with cable joint temperature and distributed temperature field data. Combined with historical health baselines, it accurately judges the equipment degradation trend. Through multi-dimensional information fusion, it clarifies the physical coordinates of the fault point, the fault type, and the confirmation conclusion, and generates accurate classification and early warning information. This enables early prediction and accurate location of faults in high-voltage power equipment, ensuring equipment operation stability, reducing maintenance costs, and improving the safety operation level of the power system. Attached Figure Description

[0059] Figure 1 This is a system architecture diagram of a power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis provided in an embodiment of the present invention;

[0060] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments belong to some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “said” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0063] Depending on the context, the word "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0064] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0065] In practice, the server-side equipment deployed in the power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis may consist of one or more devices. This system can be implemented as a business instance, a virtual machine, or hardware devices. For example, it can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, it can be understood as software deployed on a cloud node to provide monitoring and early warning services to various user terminals. Alternatively, it can be implemented as a virtual machine deployed on one or more devices in a cloud node, with application software installed to manage these user terminals. Or, it can also be implemented as a server composed of numerous identical or different types of hardware devices, with one or more hardware devices configured to provide monitoring and early warning services to various user terminals.

[0066] In terms of implementation, the monitoring and early warning system for high-voltage power equipment based on multi-parameter fusion analysis and the user terminal are mutually compatible. That is, if the monitoring and early warning system for high-voltage power equipment based on multi-parameter fusion analysis is implemented as an application installed on a cloud service platform, then the user terminal is a client that establishes a communication connection with the application; or if the monitoring and early warning system for high-voltage power equipment based on multi-parameter fusion analysis is implemented as a website, then the user terminal is implemented as a webpage; or if the monitoring and early warning system for high-voltage power equipment based on multi-parameter fusion analysis is implemented as a cloud service platform, then the user terminal is implemented as a mini-program in an instant messaging application.

[0067] like Figure 1 The figure shown is a system architecture diagram of a power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis provided by an embodiment of the present invention.

[0068] The power high-voltage equipment monitoring and early warning system 100 based on multi-parameter fusion analysis described in this invention can be installed on a cloud server. In terms of implementation, it can be used as one or more service devices, or as an application installed on the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed as a website. Depending on the functions implemented, the power high-voltage equipment monitoring and early warning system 100 based on multi-parameter fusion analysis may include a synchronization calibration module 101, a feature stripping module 102, a coupling verification module 103, a temperature field correlation module 104, a trend analysis module 105, and a fusion early warning module 106. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device's processor and can perform a fixed function, stored in the electronic device's memory.

[0069] In this embodiment of the invention, in the power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis, each of the above modules can be implemented independently and can be called by other modules. Here, "calling" can be understood as one module connecting to multiple modules of another type and providing corresponding services to those connected modules. In the power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis provided by this embodiment of the invention, the applicability of the system architecture can be adjusted by adding modules and directly calling them without modifying the program code, achieving cluster-based horizontal expansion to quickly and flexibly expand the power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis. In practical applications, the above modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in a cloud server.

[0070] The following describes, with reference to specific embodiments, each component and its specific workflow of a power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis:

[0071] The synchronization calibration module 101 is used to perform timestamp calibration on the multi-source heterogeneous data of the high-voltage power equipment to obtain the synchronous heterogeneous dataset of the high-voltage power equipment.

[0072] In this embodiment of the invention, when the synchronization calibration module performs timestamp calibration on the multi-source heterogeneous data of the high-voltage power equipment to obtain the synchronous heterogeneous dataset of the high-voltage power equipment, it is specifically used for:

[0073] The system synchronously receives distributed temperature field data of cable lines, cable joint temperature data, high-frequency current signals and traveling wave signals, and contact temperature data of switchgear, infrared thermal imaging data and transient ground voltage signals from the high-voltage power equipment to obtain a multi-source heterogeneous dataset of the high-voltage power equipment.

[0074] Time synchronization and alignment are performed on the multi-source heterogeneous dataset to obtain the multi-source signal set of the high-voltage power equipment;

[0075] The multi-source signal set is serialized and encapsulated to obtain the synchronous heterogeneous data of the high-voltage power equipment.

[0076] By deploying dedicated sensors and detection devices on high-voltage power equipment, distributed temperature field data and cable joint temperature data of cable lines are collected in real time. High-frequency current signals are captured by current detection elements, traveling wave signals are acquired by traveling wave monitoring equipment, contact temperature data are collected by temperature acquisition components on switchgear, infrared thermal imaging data is acquired by infrared thermal imaging equipment, and transient ground voltage signals are collected by transient ground voltage detection sensors. All these different types of data are collected and aggregated simultaneously through a unified signal transmission link to form a multi-source heterogeneous dataset of high-voltage power equipment.

[0077] Using a high-precision time benchmark as a reference, the acquisition time information of various types of data in the multi-source heterogeneous dataset is extracted one by one. The differences in time records of different acquisition devices are compared, and the time deviation of various types of data is corrected through a time calibration mechanism so that all data are accurately mapped to the same time dimension. This ensures that the operating data of various devices at the same time node can be accurately matched. After such time calibration processing, a multi-source signal set of high-voltage power equipment is obtained.

[0078] The data from the multi-source signal set are sorted and arranged in an orderly manner according to the data type and attributes. Each type of data is given a unique source identifier and attribute description to clearly distinguish the corresponding equipment location and monitoring type. Then, standardized encapsulation specifications are used to integrate these orderly arranged and clearly identified data into a complete data unit to ensure that the data can be accurately identified and retrieved in subsequent processing, and finally obtain the synchronous heterogeneous data of high-voltage power equipment.

[0079] The beneficial effects include the simultaneous reception of various types of monitoring data from different parts of high-voltage power equipment, ensuring the comprehensiveness and integrity of data acquisition, eliminating time deviations of various data through time synchronization and alignment, ensuring the consistency and correlation of data from different sources in the time dimension, and then serializing and encapsulating the data to form a standardized and unified structure, which facilitates accurate calling and efficient processing by subsequent modules. This provides a reliable and standardized data foundation for interference removal, feature extraction, and in-depth analysis of various data for subsequent high-frequency current signals and transient ground voltage signals, helping to improve the overall accuracy and efficiency of monitoring and early warning of high-voltage power equipment.

[0080] The stripping feature module 102 is used to strip the high-frequency current signal and transient ground voltage signal in the synchronous heterogeneous dataset to obtain the phase amplitude spectrum feature and pulse group density energy distribution feature of the power high voltage equipment.

[0081] In this embodiment of the invention, when the feature stripping module performs synchronization interference stripping on the high-frequency current signal and transient ground voltage signal in the synchronous heterogeneous dataset to obtain the phase amplitude spectrum features and pulse group density energy distribution features of the high-voltage power equipment, it is specifically used for:

[0082] The high-frequency current data in the synchronous heterogeneous dataset is subjected to pulse waveform discrimination to obtain the high-frequency current pulse signal of the high-voltage power equipment.

[0083] Pulse event detection is performed on the transient ground voltage data in the synchronous heterogeneous dataset to obtain the transient ground voltage pulse event sequence of the power high-voltage equipment;

[0084] The phase position and current amplitude of the discharge pulse in the high-frequency current signal are correlated and matrixed to construct the phase amplitude spectrum characteristics of the high-voltage power equipment.

[0085] Based on the spatiotemporal clustering characteristics and energy fluctuation characteristics of sudden discharges in the high-voltage power equipment, a differentiated characteristic criterion for the high-voltage power equipment is constructed.

[0086] Based on the aforementioned differential characteristic criteria, clustered energy density statistics are performed on the transient ground voltage pulse event sequence to obtain the pulse group density energy distribution characteristics of the high-voltage power equipment.

[0087] When the stripping feature module performs clustered energy density statistics on the transient ground voltage pulse event sequence based on the differentiated feature criteria to obtain the pulse group density energy distribution characteristics of the high-voltage power equipment, it is specifically used for:

[0088] Based on the spatiotemporal cluster characteristics, the transient ground voltage pulse event sequence is divided into group boundaries to obtain the sudden pulse group of the power high-voltage equipment;

[0089] Time-domain clustering analysis was performed on the burst pulse groups to obtain the interval distribution and duration distribution of the high-voltage power equipment.

[0090] Evolution pattern analysis of the energy intensity of the sudden pulse group yields the pulse energy intensity distribution and evolution trajectory characteristics of the high-voltage power equipment.

[0091] The pulse group density energy distribution characteristics of the high-voltage power equipment are obtained by cross-latitude fusion of the interval distribution, the duration distribution, the pulse energy intensity distribution, and the evolution trajectory characteristics.

[0092] When the stripping feature module performs evolution pattern analysis on the energy intensity of the sudden pulse group to obtain the pulse energy intensity distribution and evolution trajectory characteristics of the high-voltage power equipment, it is specifically used for:

[0093] Based on the amplitude information of the pulse events in the burst pulse group, energy distribution analysis is performed on the pulse events to obtain the pulse energy intensity distribution of the high-voltage power equipment;

[0094] By analyzing the statistical distribution law of the energy release concentration and discrete characteristics of the pulse energy intensity sequence, a description of the intensity distribution of the high-voltage power equipment is obtained.

[0095] Based on the pulse energy intensity sequence, the energy gradient change of the sudden pulse group is identified to obtain a description of the dynamic evolution trend of the high-voltage power equipment.

[0096] Semantic trajectory evolution is performed on the intensity distribution description and the dynamic evolution trend description to obtain the evolution trajectory characteristics of high-voltage power equipment.

[0097] By analyzing the waveform morphology of high-frequency current data in synchronous heterogeneous datasets and comparing it with the reference waveform of high-frequency current during normal operation of high-voltage power equipment, the focus is on the amplitude variation law, the steepness of the rising and falling edges, and the duration characteristics of the waveform. Waveforms consistent with the characteristics of standard discharge pulse waveforms are retained, while irregular interference waveforms caused by grid interference, equipment vibration, etc. are eliminated, thereby obtaining the high-frequency current pulse signal of high-voltage power equipment.

[0098] The normal fluctuation range of transient ground voltage of high-voltage power equipment is set as a reference benchmark. The transient ground voltage data in the synchronous heterogeneous dataset is continuously monitored. When the data exceeds the reference benchmark and exhibits instantaneous change characteristics, it is determined to be a transient ground voltage pulse event. According to the chronological order of the events, all detected pulse events are arranged in sequence to form a transient ground voltage pulse event sequence of high-voltage power equipment.

[0099] The phase information corresponding to each discharge pulse in the high-frequency current pulse signal is extracted to determine the phase position. At the same time, the current amplitude data of each discharge pulse is recorded. A two-dimensional correlation framework is constructed with the phase position as the horizontal coordinate dimension and the current amplitude as the vertical coordinate dimension. The matching relationship between each phase position and the corresponding current amplitude is filled into the framework one by one to form a matrix structure that can intuitively reflect the correlation between the two. Based on this matrix structure, the phase amplitude spectrum feature of the high-voltage power equipment is generated.

[0100] This study delves into the concentrated occurrence patterns of sudden discharges in high-voltage power equipment over time and their specific regional distribution characteristics in space, clarifying the core manifestations of spatiotemporal clustering. It also records the alternating strengths and weaknesses of energy release during discharge, outlining the specific patterns of energy fluctuations. Combining these two characteristics, it develops criteria to distinguish sudden discharge pulses from other interference pulses, defines the spatiotemporal distribution conditions and energy change thresholds that effective pulses must meet, and constructs differentiated characteristic criteria for high-voltage power equipment.

[0101] Based on the differentiation criterion, each pulse event in the transient ground voltage pulse event sequence is screened, and valid pulse events that meet the criteria are retained. These valid pulse events are divided into multiple pulse groups according to time interval or spatial correlation. The number of pulse events, the energy value of a single pulse, and the duration of the group are counted in each group. The total energy and pulse density of each group per unit time are calculated. The energy density data of each group and the distribution differences between different groups are integrated to obtain the pulse group density energy distribution characteristics of the high-voltage power equipment.

[0102] Based on the spatiotemporal clustering characteristics of sudden discharges, this study focuses on observing the temporal occurrence patterns and spatial correlation attributes of pulse events in transient ground voltage pulse event sequences. The temporal closeness of consecutive pulse events is used as the basis for time division. When the time interval between adjacent pulse events is within a set close interval, they are determined to be consecutive events in the same group. At the same time, combined with the physical area of ​​the high-voltage power equipment corresponding to the pulse events, pulse events in the same equipment area are grouped into the same spatial correlation group. Through the definition of time and space, the start and end nodes of different groups are clearly distinguished, the group boundary is delineated, and the sudden pulse groups of high-voltage power equipment are obtained.

[0103] For each burst pulse cluster, the specific time point of each pulse event within the cluster is recorded one by one, the time interval between two adjacent pulse events is calculated, and all interval data are sorted in order of magnitude to form an interval distribution that can reflect the time distribution pattern of pulse events within the cluster. At the same time, starting from the time of the first pulse event in each burst pulse cluster, the duration of the entire process is recorded until the time of the last pulse event in the cluster ends. The duration data of all burst pulse clusters are collected, classified and statistically analyzed to obtain the duration distribution of high-voltage power equipment.

[0104] The amplitude information of each pulse event in the burst pulse cluster is extracted, and the energy intensity value of each pulse event is obtained by converting the amplitude to energy intensity according to the correspondence between amplitude and energy intensity. The distribution of these energy intensity values ​​in different intensity ranges is counted to clarify the proportion of pulses of various energy intensities, and the pulse energy intensity distribution of the high-voltage power equipment is obtained. The change of energy intensity in each burst pulse cluster over time is continuously tracked, and the peak occurrence time, change amplitude, and overall change trend of energy intensity are recorded, such as whether it gradually increases, slowly decreases, or fluctuates. These energy change characteristics in the time dimension are sorted and integrated to obtain the evolution trajectory characteristics of the high-voltage power equipment.

[0105] By comprehensively integrating the pulse density distribution within the pulse group as reflected by the interval distribution, the pulse group duration distribution as reflected by the duration distribution, the energy magnitude distribution pattern shown by the pulse energy intensity distribution, and the energy change trend reflected by the evolution trajectory characteristics, a multi-dimensional approach is adopted. Each feature is an important component of the pulse group density energy distribution. By clarifying the correlation between the features, such as the matching between density and energy intensity, and the correspondence between duration and evolution trend, the dispersed features are organically integrated into a holistic feature that comprehensively reflects the pulse group density and energy distribution state, thus obtaining the pulse group density energy distribution characteristics of high-voltage power equipment.

[0106] The correspondence between pulse event amplitude and energy intensity is clarified. The amplitude information of each pulse event in the sudden pulse group is extracted one by one. Fixed interval ranges are divided according to the energy intensity. Each pulse event is assigned to the corresponding interval according to the energy intensity corresponding to its amplitude. The number of pulse events contained in each interval is counted. The distribution of pulse events under different energy intensity levels is clearly presented, and the pulse energy intensity distribution of high-voltage power equipment is obtained.

[0107] All pulse events in the pulse energy intensity distribution are arranged sequentially according to their energy intensity to form a pulse energy intensity sequence. The clustering of energy values ​​in the sequence is observed to determine the energy range in which most pulse events are concentrated, and to determine whether the energy release is concentrated in a specific interval. At the same time, the differences between the energy values ​​of different pulse events are analyzed to clarify the degree of dispersion of energy distribution. The inherent laws of these concentrated characteristics and discrete manifestations are sorted out to form a textual description of the overall characteristics of the pulse energy intensity distribution, thus obtaining a description of the intensity distribution of high-voltage power equipment.

[0108] Arrange the pulse energy intensity sequence according to the chronological order of the pulse events, compare the energy intensity values ​​of adjacent pulse events one by one, observe the energy change of the later pulse relative to the previous pulse, determine whether the energy is continuously increasing, gradually decreasing, remaining stable, or fluctuating alternately, track the overall trend of energy change in the entire sequence, clarify the change law of energy intensity over time, form a textual description of the dynamic change of pulse group energy, and obtain a description of the dynamic evolution trend of high voltage power equipment.

[0109] By combining the static characteristics such as the energy concentration range and dispersion reflected in the intensity distribution description with the dynamic characteristics such as the energy change direction and rate reflected in the dynamic evolution trend description, and linking these characteristics together according to the time sequence and energy change logic, a complete trajectory description of pulse group energy from its distribution state to its dynamic changes is constructed. This clearly shows the correlation between energy intensity distribution and time evolution, and yields the evolution trajectory characteristics of high-voltage power equipment.

[0110] The beneficial effects are as follows: By performing pulse waveform identification and pulse event detection on high-frequency current data and transient ground voltage data in synchronous heterogeneous datasets, interference components are effectively separated and effective signals are screened out, obtaining pure high-frequency current pulse signals and ordered transient ground voltage pulse event sequences, laying a reliable foundation for subsequent feature extraction; by constructing a correlation matrix between the phase position of the discharge pulse and the current amplitude, a phase amplitude spectrum feature that intuitively reflects the correlation between the two is formed, helping to accurately capture the key electrical characteristics of partial discharge; based on the spatiotemporal cluster characteristics and energy fluctuation characteristics of sudden discharge, differentiated feature criteria are constructed, providing a clear basis for effectively distinguishing discharge pulses from interference pulses; through group boundary division, time-domain clustering analysis, energy intensity evolution pattern identification, and cross-dimensional fusion, the interval distribution, duration distribution, pulse energy intensity distribution, and evolution trajectory characteristics of sudden pulse groups are comprehensively explored, integrating them to form comprehensive and accurate pulse group density energy distribution characteristics, improving the effectiveness of partial discharge signal processing and the accuracy of feature extraction in high-voltage power equipment, providing comprehensive and reliable feature support for equipment insulation status assessment, fault identification, and early warning, and ensuring the accuracy and effectiveness of high-voltage power equipment operation monitoring.

[0111] The coupling verification module 103 is used to perform spatiotemporal coupling cross-verification of the phase amplitude spectrum features and the pulse group density energy distribution features to obtain the effective partial discharge events of the high-voltage power equipment.

[0112] In this embodiment of the invention, when the coupling verification module performs spatiotemporal coupling cross-verification of the phase amplitude spectrum features and the pulse group density energy distribution features to obtain the effective partial discharge event of the high-voltage power equipment, it is specifically used for:

[0113] The overlap degree of the active phase intervals of the phase amplitude spectrum features and the high-density pulse group periods of the pulse group density energy distribution features is determined to obtain the time overlap result of the high-voltage power equipment.

[0114] By performing trend analysis on the phase amplitude spectrum features and the pulse group density energy distribution features, the amplitude evolution sequence and intensity evolution sequence of the high-voltage power equipment are obtained.

[0115] Based on the time coincidence results, the trend consistency of the amplitude evolution sequence and the intensity evolution sequence is judged to obtain the trend comparison conclusion of the high-voltage power equipment.

[0116] Based on the physical coupling relationship criteria of the high-voltage power equipment, the event determination is performed on the time coincidence results and the trend comparison conclusion to obtain the effective partial discharge events of the high-voltage power equipment.

[0117] The phase range where discharge pulses are concentrated and the amplitude is at a high level is identified from the phase amplitude spectrum characteristics. This range is defined as the active phase interval. At the same time, the time range where the number of pulses is dense and the energy is concentrated is screened from the pulse group density energy distribution characteristics. This time range is defined as the high-density pulse group period. By comparing the time span corresponding to the active phase interval with the time coverage of the high-density pulse group period, the overlap between the two in the time dimension and the degree of overlap are identified, thus forming the time coincidence result of the high-voltage power equipment.

[0118] Based on the phase amplitude spectrum characteristics, discharge pulse amplitude data corresponding to each time node is extracted one by one according to the time progression. These amplitude data arranged in time order are organized into a continuous sequence to obtain the amplitude evolution sequence of the high-voltage power equipment. Based on the pulse group density energy distribution characteristics, pulse group energy intensity data corresponding to each time node is extracted in the same time order. These intensity data are arranged in chronological order to form the intensity evolution sequence of the high-voltage power equipment.

[0119] Based on the time overlap results, the focus is on the time range where the active phase interval and the high-density pulse group period overlap. The direction of change of the data in the amplitude evolution sequence within this range is compared with the direction of change of the data in the intensity evolution sequence. It is observed whether the intensity increases synchronously when the amplitude increases and whether the intensity decreases accordingly when the amplitude decreases. It is judged whether the rhythm and pattern of the two changes are consistent. Based on this matching of direction and pattern, the trend comparison conclusion of the high-voltage power equipment is drawn.

[0120] The physical coupling relationship criterion for high-voltage power equipment is based on the physical characteristics of partial discharge within the equipment. It clarifies the necessary correlation between phase amplitude changes and pulse group energy density changes when partial discharge occurs. Based on this criterion, it is determined whether the time coincidence result meets the time correlation standard corresponding to partial discharge. At the same time, it is checked whether the trend comparison conclusion conforms to the change law under the physical coupling relationship. When both the time coincidence result and the trend comparison conclusion meet the requirements of the physical coupling relationship criterion, the corresponding event is determined to be a valid partial discharge event of the high-voltage power equipment.

[0121] The beneficial effects are as follows: by determining the overlap between the active phase intervals of the phase amplitude spectrum characteristics and the high-density pulse burst periods of the pulse burst density energy distribution characteristics, the correlation between the two in the time dimension can be accurately identified, providing a reliable time basis for subsequent verification; trend analysis of the two types of features is performed to fully extract the evolution law of amplitude and intensity, forming a sequence data that comprehensively reflects the changes in equipment status; trend consistency discrimination is carried out based on the time overlap results to ensure that the change laws of the two types of features match each other, further filtering effective correlation information; event judgment is performed in combination with the physical coupling relationship criteria of high-voltage power equipment, strictly following the physical characteristics of partial discharge inside the equipment, effectively eliminating false signals and interference events, accurately identifying effective partial discharge events of high-voltage power equipment, providing real and reliable core data support for subsequent temperature field correlation analysis and fault judgment, and ensuring the accuracy and effectiveness of high-voltage power equipment status monitoring.

[0122] The temperature field correlation module 104 is used to perform multidimensional correlation analysis on the cable joint temperature data and distributed temperature field data in the effective partial discharge event based on the electrothermal coupling mechanism of insulation partial discharge in the high-voltage power equipment, so as to obtain the primary fault analysis results of the high-voltage power equipment.

[0123] In this embodiment of the invention, when the temperature field correlation module performs multidimensional correlation analysis on the cable joint temperature data and distributed temperature field data in the effective partial discharge event based on the electrothermal coupling mechanism of insulation partial discharge in high-voltage power equipment, and obtains the primary fault analysis results of the high-voltage power equipment, it is specifically used for:

[0124] Information analysis is performed on the occurrence time and signal source attributes of the effective partial discharge events to obtain the target time and associated spatial location of temperature analysis in the high-voltage power equipment;

[0125] Based on the target time, the rate of change of cable joint temperature data in the synchronous heterogeneous dataset is identified to obtain the transient rise rate of the high-voltage power equipment.

[0126] Based on the associated spatial location, temperature field distribution analysis is performed on the distributed temperature field data in the synchronous heterogeneous dataset to obtain the local thermal focusing phenomenon of the high-voltage power equipment.

[0127] Multidimensional spatiotemporal synergy verification was performed on the effective partial discharge event, the transient rise rate, the local thermal focusing phenomenon, and the abnormal disturbance of the traveling wave time difference signal in the synchronous heterogeneous data set to obtain the primary fault analysis results of the high-voltage power equipment.

[0128] The specific occurrence time is directly extracted from the recorded information of effective partial discharge events. At the same time, by analyzing the signal propagation path, amplitude attenuation law and frequency characteristics corresponding to the event, the specific components and areas of the high-voltage power equipment corresponding to the signal source are determined. The extracted occurrence time and the related time periods before and after are defined as the target time for temperature analysis. The equipment components and areas corresponding to the signal source are determined as the associated spatial locations for temperature analysis, thus obtaining the target time and associated spatial locations for temperature analysis in high-voltage power equipment.

[0129] Cable joint temperature data corresponding to associated spatial locations are filtered from synchronous heterogeneous datasets. Focusing on temperature records within the target time range, the cable joint temperature value at the target time point and the temperature value of the adjacent stable period before the target time are extracted. By comparing the difference between these two temperature values ​​and combining the interval between the two time points, the change range of cable joint temperature within a unit time is determined, and the transient rise rate of the high-voltage power equipment is obtained.

[0130] Based on the associated spatial location, the corresponding distributed temperature field data range in the synchronous heterogeneous dataset is located. The temperature data within this range is checked point by point and divided into regions. By comparing the temperature values ​​of different regions, regions with significantly higher temperature values ​​than the surrounding regions and exhibiting a concentrated distribution are identified. The temperature distribution pattern and range of these regions are observed to clarify the core area of ​​temperature accumulation and obtain the local heat focusing phenomenon of high-voltage power equipment.

[0131] Verify whether the occurrence time of the effective partial discharge event, the occurrence time of the transient rise rate, and the formation time of the local thermal focusing phenomenon are within the same time period. Confirm whether the signal source location of the effective partial discharge event, the cable joint location corresponding to the transient rise rate, and the occurrence area of ​​the local thermal focusing phenomenon are the same part of the high-voltage power equipment or adjacent related areas. At the same time, check whether the abnormal disturbance of the traveling wave time difference signal in the synchronous heterogeneous data center also appears in the same spatiotemporal range. Determine whether these four types of factors match and corroborate each other in the time and space dimensions. When all factors meet the spatiotemporal coordination relationship, integrate them to form the primary fault analysis results of the high-voltage power equipment.

[0132] The beneficial effects include: based on the electrothermal coupling mechanism of partial discharge in insulation, by analyzing the occurrence time and signal source attributes of effective partial discharge events, the target time and associated spatial location of temperature analysis can be accurately located, providing a clear direction for subsequent targeted analysis of temperature data; based on the target time, the transient rise rate of cable joint temperature can be identified, which can keenly capture the rapid temperature changes caused by partial discharge and truly reflect the immediate effect of electrothermal coupling; based on the associated spatial location, the local heat focusing phenomenon of the distributed temperature field can be identified, which can accurately locate the temperature concentration area and comprehensively present the local thermal distribution anomalies of the equipment; through multi-dimensional spatiotemporal synergistic verification of effective partial discharge events, transient rise rates, local heat focusing phenomena, and abnormal disturbances of traveling wave time difference signals, the mutual verification and supplementation of multi-source data can be achieved, effectively eliminating the interference and misjudgment of single data, ensuring the accuracy and reliability of the primary fault analysis results, providing a comprehensive and solid data analysis foundation for subsequent accurate fault judgment and insulation status assessment of high-voltage power equipment, helping to timely discover potential hidden dangers of equipment, and ensuring the safe and stable operation of high-voltage power equipment.

[0133] The trend analysis module 105 is used to comprehensively analyze the deviation and evolution rate of the synchronous heterogeneous dataset based on the historical health baseline of the high-voltage power equipment, so as to obtain the deterioration trend conclusion of the high-voltage power equipment.

[0134] In this embodiment of the invention, when the trend analysis module performs a comprehensive analysis of the deviation and evolution rate of the synchronous heterogeneous dataset based on the historical health baseline of the high-voltage power equipment to obtain a conclusion on the deterioration trend of the high-voltage power equipment, it is specifically used for:

[0135] Based on the historical health baseline, the synchronous heterogeneous dataset is state-aligned to obtain the comparison data pair sequence of the high-voltage power equipment;

[0136] The difference degree of the comparison data sequence is described to obtain the deviation description sequence of the high-voltage power equipment;

[0137] Based on the continuous time window of the synchronous heterogeneous dataset, the direction and rate of change of the deviation description sequence are tracked and described to obtain the evolution rate description of the high-voltage power equipment;

[0138] An evolution trajectory is constructed from the deviation description sequence and the evolution rate description to obtain the degradation trajectory of the high-voltage power equipment.

[0139] The location information of the degradation trajectory is mapped to a preset stability threshold boundary to obtain the degradation trend conclusion of the high-voltage power equipment.

[0140] Historical health baselines are benchmark standards formed by long-term monitoring and accumulation of multi-source heterogeneous data of high-voltage power equipment under normal operating conditions. They cover the characteristic range and change patterns of various normal operating parameters. According to the type, monitoring dimension and time attribute of various data in the synchronous heterogeneous dataset, each data item is precisely matched with the corresponding benchmark data in the historical health baseline, so that the current data and the benchmark data correspond one-to-one to form pairs. These pairs are then organized according to the chronological order of data collection to obtain the comparison data pair sequence of high-voltage power equipment.

[0141] Extract each corresponding data set from the comparison data pair sequence one by one, and directly compare the current data in the synchronous heterogeneous dataset with the baseline data in the historical health baseline to determine whether the current data is higher, lower, or within the normal range relative to the baseline data. Describe in detail the degree of difference between the two, such as slight difference, moderate deviation, or significant deviation. Arrange the description of the degree of difference of each data set in the time order of the comparison data pair sequence to form a deviation description sequence of high voltage power equipment.

[0142] A continuous time window refers to a continuous and uninterrupted monitoring period. According to the division method of this time period, the deviation description sequence is divided into multiple continuous segments. The changes in the deviation description are analyzed segment by segment. It is observed whether the deviation in each segment gradually increases, gradually decreases, or remains stable. At the same time, the speed of this change is judged, such as rapid change or slow change. By combining the direction and speed of change in each continuous time window, a complete description of the deviation change characteristics is formed, and the evolution rate description of the high-voltage power equipment is obtained.

[0143] Using time as a connecting thread, the degree of difference at each time point in the deviation description sequence is combined with the speed and direction of change at different stages in the evolution rate description. Following the order of time progression, these scattered descriptive information are organically linked together, clearly showing the entire process of how the degree of deviation of the high-voltage power equipment from the normal benchmark changes over time, starting from a certain state. This includes the initial state of the difference, the intermediate evolution process, and the current state, constructing an overall trajectory that can intuitively reflect the path of equipment state change, and obtaining the degradation trajectory of the high-voltage power equipment.

[0144] The preset stability threshold boundary is determined based on the design standards, operating requirements, and safety specifications of high-voltage power equipment. It defines the range of stable operation of the equipment and clarifies that exceeding this range may lead to failure risks. The positions of each state corresponding to the constructed degradation trajectory are compared one by one with the stability threshold boundary to determine whether the degradation trajectory is in a stable region within the boundary, a warning region near the boundary, or a risk region beyond the boundary. At the same time, combined with the trend of trajectory changes, it is determined whether the equipment status is continuously deteriorating, remaining stable, or trending towards improvement, and finally, a conclusion on the degradation trend of the high-voltage power equipment is obtained.

[0145] The beneficial effects are as follows: Based on the historical health baseline of high-voltage power equipment, the synchronous heterogeneous dataset is aligned to construct an accurate comparison data pair sequence, providing a reliable benchmark for equipment condition difference analysis; by describing the degree of difference in the comparison data pair sequence, the deviation of the current equipment condition from the health benchmark is clearly presented, forming a comprehensive deviation description sequence; relying on continuous time windows to track the direction and rate of change of the deviation description sequence, the dynamic characteristics of equipment condition changes are accurately captured, generating a detailed evolution rate description; combining the deviation description sequence with the evolution rate description constructs a degradation trajectory, intuitively showing the evolution path of the equipment condition; by mapping the degradation trajectory to a preset stability threshold boundary, the stability level and evolution trend of the equipment condition are clarified, and accurate degradation trend conclusions are obtained, providing a scientific basis for preventive maintenance of high-voltage power equipment, helping to avoid fault risks in advance, and ensuring the long-term stable operation of the equipment.

[0146] The fusion early warning module 106 is used to perform multi-dimensional information fusion on the primary fault analysis results and the deterioration trend conclusions to obtain the classification early warning information of the high-voltage power equipment.

[0147] In this embodiment of the invention, when the fusion early warning module performs multi-dimensional information fusion on the primary fault analysis results and the degradation trend conclusions to obtain the classification early warning information of the high-voltage power equipment, it is specifically used for:

[0148] By jointly determining the results of the initial fault analysis and the conclusion of the deterioration trend, a fault confirmation conclusion is obtained for the high-voltage power equipment.

[0149] Based on the traveling wave time difference signal in the synchronous heterogeneous dataset, the fault section of the cable line with abnormal fault in the high-voltage power equipment is defined to obtain the preliminary fault section range of the high-voltage power equipment.

[0150] Based on the preliminary fault zone range, multi-source spatial information fusion is performed on the spatial center of the local thermal focusing phenomenon in the high-voltage power equipment, the specific location of the abnormal temperature node, and the location of the dominant signal source of the effective partial discharge event to obtain the physical coordinates of the fault point of the high-voltage power equipment.

[0151] Based on the dominant abnormal signal characteristics in the primary fault analysis results, the nature of the abnormal fault is determined to obtain the fault type of the high-voltage power equipment.

[0152] The fault confirmation conclusion, the physical coordinates of the fault point, and the fault type are encapsulated to obtain the classification and early warning information of the high-voltage power equipment.

[0153] The formula for calculating the physical coordinates of the fault point is as follows:

[0154] ;

[0155] In the formula, The physical coordinates of the fault point are: The coordinates of the spatial center of the localized thermal focusing phenomenon are: These are the specific location coordinates of the abnormal temperature node. The coordinates are the location coordinates of the dominant signal source of the effective partial discharge event. Assign a preset confidence weight to the thermal focusing location. Assign a preset confidence weight to the temperature node location. The preset confidence weights for the location of the discharge signal source are used.

[0156] The abnormal phenomena, potential fault locations, and related data characteristics reflected in the primary fault analysis results are comprehensively compared with the equipment state evolution direction, degradation rate, and stability level reflected in the degradation trend conclusion. The consistency between the two in terms of fault direction, scope of impact, and severity is verified. When the anomalies confirmed by the primary fault analysis results and the continuous degradation state presented by the degradation trend conclusion corroborate each other, it is clearly determined that there is a real fault in the equipment. At the same time, the severity level of the fault is determined by combining the information from both, thus forming a fault confirmation conclusion for the high-voltage power equipment.

[0157] The traveling wave time difference signal of the cable line is extracted from the synchronous heterogeneous dataset. The propagation time difference of the traveling wave between different monitoring points of the cable line is analyzed. Based on the correspondence between the traveling wave propagation speed and the line length, combined with the distribution of the monitoring points, the initial propagation section of the abnormal traveling wave disturbance is identified. By tracking the change trajectory of the traveling wave time difference signal, normal fluctuation interference is eliminated, and the range of the cable line where the abnormal fault is located is clearly defined, thus obtaining the preliminary fault section range of the high-voltage power equipment.

[0158] Using the initial fault zone as a defined area, the spatial center of the local thermal focusing phenomenon within this area is extracted, the specific distribution points of the abnormal temperature nodes are determined, and the location of the dominant signal source of the effective partial discharge event is located. The spatial correlation and overlap of these three locations are analyzed. Based on the physical laws at the time of the fault, the directional weight of each location to the fault is comprehensively considered, and the information of the three locations is spatially superimposed and integrated to accurately determine the specific location that reflects the core source of the fault, thus obtaining the physical coordinates of the fault point of the high-voltage power equipment.

[0159] The dominant abnormal signal characteristics in the primary fault analysis results are analyzed in depth, including the waveform shape, frequency distribution, energy intensity and duration of the signal. These characteristics are matched one by one with the typical signal characteristics corresponding to common fault types of high-voltage power equipment, such as the signal attenuation law corresponding to insulation aging and the signal fluctuation characteristics corresponding to poor contact. Through comprehensive comparison and accurate identification of the characteristics, the specific category to which the current abnormal fault belongs is determined, and the fault type of the high-voltage power equipment is obtained.

[0160] The information regarding the fault's existence status and severity level, the specific location information corresponding to the physical coordinates of the fault point, and the specific fault category information corresponding to the fault type in the fault confirmation conclusion are organized and classified according to a unified information standard. Clear labels and descriptions are added to each type of information to ensure that all types of information are clear, easy to identify, and integrated into a complete early warning data unit through a standardized encapsulation method, forming classified early warning information for high-voltage power equipment that can accurately guide equipment maintenance.

[0161] The spatial center coordinates of the localized heat focusing phenomenon are derived from the analysis of the distributed temperature field data in the synchronous heterogeneous dataset by the temperature field association module. By locking down the core area within the associated spatial location where the temperature is significantly higher than the surrounding area and is concentrated, the coordinate information corresponding to the spatial center of this area is determined.

[0162] The specific location coordinates of abnormal temperature nodes are obtained from the analysis of centralized cable joint temperature data and distributed temperature field data in the synchronous heterogeneous dataset by the temperature field association module. Within the associated spatial location range, the specific locations of temperature anomalies are investigated, and the spatial coordinates corresponding to these abnormal locations are determined.

[0163] The location coordinates of the dominant signal source of the effective partial discharge event are obtained from the effective partial discharge event obtained by the coupling verification module. By analyzing the signal propagation path, amplitude attenuation law and frequency characteristics of the event, the specific spatial coordinates of the high-voltage power equipment corresponding to the signal source are determined.

[0164] The confidence weights for thermal focusing location, temperature node location, and discharge signal source location are fixed weights pre-set based on the design standards, operating characteristics, and historical fault data of high-voltage power equipment, combined with different location information to reflect the importance of various location information in fault location.

[0165] The significance of this formula lies in comprehensively integrating three types of fault-related spatial information: the spatial center of the local thermal focusing phenomenon, the specific location of the abnormal temperature node, and the location of the dominant signal source of the effective partial discharge event. By balancing the influence of various types of location information on fault location through preset confidence weights, it avoids the possible deviation of single location information, accurately calculates the physical coordinates that can truly reflect the core source of the fault in high-voltage power equipment, provides a scientific basis for accurate fault location, and ensures that subsequent maintenance work can be carried out directly on the fault point.

[0166] The beneficial effects include: achieving accurate fault confirmation by jointly judging the results of primary fault analysis and the conclusions of degradation trends, avoiding misjudgments caused by single data, and ensuring the authenticity and reliability of fault conclusions; defining the cable line section with abnormal faults based on traveling wave time difference signals, quickly locating the fault range, and providing a clear basis for subsequent precise positioning; accurately determining the physical coordinates of the fault point by integrating multi-source spatial information such as the spatial center of local thermal focusing phenomena, the location of abnormal temperature nodes, and the location of the dominant signal source of effective partial discharge events, achieving precise fault location; identifying the fault type based on the dominant abnormal signal characteristics in the primary fault analysis results, clearly defining the specific nature of the fault, and providing key references for targeted treatment; and encapsulating the fault confirmation conclusions, the physical coordinates of the fault point, and the fault type to form standardized and unified classification and early warning information, enabling relevant personnel to quickly obtain core fault information, efficiently carry out equipment maintenance work, effectively reduce fault handling time and costs, and ensure the safe and stable operation of high-voltage power equipment.

[0167] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0168] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0169] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis, characterized in that, The system includes a synchronous calibration module, a feature stripping module, a coupling verification module, a temperature field correlation module, a trend analysis module, and a fusion early warning module, wherein: The synchronization calibration module timestamps the multi-source heterogeneous data from high-voltage power equipment to obtain a synchronous heterogeneous dataset. The stripping feature module performs synchronous interference stripping on the high-frequency current signal and transient ground voltage signal in the synchronous heterogeneous dataset to obtain phase amplitude spectrum features and pulse group energy density distribution features. The coupling verification module performs time overlap determination on the active phase interval of the phase amplitude spectrum feature and the high-density pulse group period of the pulse group energy density distribution feature to obtain time overlap results; it performs trend analysis on the phase amplitude spectrum feature and the pulse group energy density distribution feature to obtain the amplitude evolution sequence of the discharge pulse and the intensity evolution sequence of the pulse group energy; based on the time overlap results, it performs trend consistency determination on the amplitude evolution sequence and the intensity evolution sequence to obtain trend comparison conclusions; when both the time overlap results and the trend comparison conclusions meet the requirements of the physical coupling relationship criterion, the corresponding event is determined as a valid partial discharge event of the high-voltage power equipment; the physical coupling relationship criterion is based on the physical characteristics of partial discharge inside the equipment, clarifying the necessary correlation between the phase amplitude change and the pulse group energy density change when partial discharge occurs, thereby determining whether the time overlap results meet the time correlation standard corresponding to partial discharge, and simultaneously verifying whether the trend comparison conclusions conform to the change law under the physical coupling relationship; The temperature field correlation module parses the occurrence time and signal source attributes of the effective partial discharge event to obtain the target time and associated spatial location for temperature analysis in the high-voltage power equipment. Based on the target time, it identifies the rate of change of cable joint temperature data in the synchronous heterogeneous dataset to obtain the transient temperature rise rate. Based on the associated spatial location, it analyzes the temperature field distribution of distributed temperature field data in the synchronous heterogeneous dataset to obtain the local heat focusing phenomenon of the high-voltage power equipment. It then determines whether the effective partial discharge event, the transient temperature rise rate, the local heat focusing phenomenon, and the abnormal disturbance of the traveling wave time difference signal in the synchronous heterogeneous dataset match and corroborate each other in time and space to obtain the primary fault analysis results of the high-voltage power equipment. The trend analysis module, based on the historical health baseline of the high-voltage power equipment, comprehensively analyzes the deviation and evolution rate of the synchronous heterogeneous dataset to obtain a conclusion on the deterioration trend of the high-voltage power equipment. The fusion early warning module performs multi-dimensional information fusion on the primary fault analysis results and the deterioration trend conclusions to obtain the classification early warning information of the high-voltage power equipment.

2. The power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis according to claim 1, characterized in that, When the synchronization calibration module performs timestamp calibration on multi-source heterogeneous data from high-voltage power equipment to obtain a synchronized heterogeneous dataset, it is specifically used for: The system simultaneously receives distributed temperature field data, cable joint temperature data, high-frequency current signals and traveling wave signals from cable lines in high-voltage power equipment, as well as contact temperature data, infrared thermal imaging data and transient ground voltage signals from switchgear, to obtain multi-source heterogeneous data of the high-voltage power equipment. The multi-source heterogeneous data is time-synchronized and aligned to obtain a multi-source signal set; The multi-source signal set is serialized and encapsulated to obtain the synchronous heterogeneous dataset. 3.The power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis of claim 1, wherein, When the stripping feature module performs synchronization interference stripping on the high-frequency current signal and transient ground voltage signal in the synchronous heterogeneous dataset to obtain phase amplitude spectrum features and pulse group energy density distribution features, it is specifically used for: The high-frequency current signal in the synchronous heterogeneous dataset is subjected to pulse waveform discrimination to obtain the high-frequency current pulse signal; Pulse event detection is performed on the transient ground voltage signal in the synchronous heterogeneous dataset to obtain a transient voltage pulse event sequence; The phase position and current amplitude of each discharge pulse in the high-frequency current pulse signal are correlated and matrixed to obtain the phase amplitude spectrum features; Based on the spatiotemporal clustering characteristics and energy fluctuation characteristics of sudden discharges in the high-voltage power equipment, a judgment criterion for distinguishing sudden discharge pulses from other interference pulses is formulated, the spatiotemporal distribution conditions and energy change thresholds that effective pulses must meet are determined, and a differentiated characteristic criterion is constructed. Based on the aforementioned differential characteristic criteria, clustered energy density statistics are performed on the transient ground voltage pulse event sequence to obtain the pulse group energy density distribution characteristics.

4. The power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis as described in claim 3, characterized in that, When the stripping feature module performs clustered energy density statistics on the transient ground voltage pulse event sequence based on the differentiated feature criteria to obtain the pulse group energy density distribution characteristics, it is specifically used for: Based on the spatiotemporal clustering characteristics, the transient ground voltage pulse event sequence is divided into group boundaries to obtain a burst pulse group; A time-domain clustering analysis was performed on the sudden burst of pulses to calculate the time interval between two adjacent pulse events. All interval data were sorted in order of size to form an interval distribution that can reflect the temporal distribution pattern of pulse events within the cluster. Simultaneously, starting from the time of the first pulse event in each burst burst, the duration of the entire process is recorded until the time of the last pulse event in the burst burst ends. The duration data of all burst bursts are collected, classified, and statistically analyzed to obtain the duration distribution. The energy intensity of the sudden pulse group is analyzed to determine the evolution pattern, and the pulse energy intensity distribution and evolution trajectory characteristics are obtained. The pulse group energy density distribution characteristics are obtained by cross-latitude fusion of the interval distribution, the duration distribution, the pulse energy intensity distribution, and the evolution trajectory characteristics.

5. The power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis as described in claim 4, characterized in that, When the stripping feature module performs evolution pattern analysis on the energy intensity of the sudden pulse cluster to obtain pulse energy intensity distribution and evolution trajectory features, it is specifically used for: Based on the amplitude information of the pulse events in the burst pulse group, energy distribution analysis is performed on the pulse events to obtain the pulse energy intensity distribution; The intensity distribution description is obtained by statistically analyzing the concentration and discreteness of energy release in the pulse energy intensity distribution. Based on the pulse energy intensity distribution, the energy gradient change of the sudden pulse group is identified to obtain a description of the dynamic evolution trend. Semantic trajectory evolution is performed on the intensity distribution description and the dynamic evolution trend description to obtain evolution trajectory features.

6. The power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis as described in claim 1, characterized in that, When the trend analysis module performs a comprehensive analysis of the deviation and evolution rate of the synchronous heterogeneous dataset based on the historical health baseline of the high-voltage power equipment to obtain a conclusion on the deterioration trend of the high-voltage power equipment, it is specifically used for: According to the data types, monitoring dimensions and time attributes of various data in the synchronous heterogeneous dataset, each data item is matched with the corresponding benchmark data in the historical health baseline, so that the current data and the benchmark data correspond one-to-one to form a pair. These pairs are arranged in the order of data collection time to obtain the comparison data pair sequence. The alignment data is used to describe the degree of difference between the sequences, resulting in a deviation description sequence. Based on the continuous time window of the synchronous heterogeneous dataset, the direction and rate of change of the deviation description sequence are tracked and described to obtain the evolution rate description; An evolution trajectory is constructed from the deviation description sequence and the evolution rate description to obtain the degradation trajectory of the high-voltage power equipment. The location information of the degradation trajectory is mapped to a preset stability threshold boundary to obtain the degradation trend conclusion of the high-voltage power equipment.

7. The power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis as described in claim 1, characterized in that, When the fusion early warning module performs multi-dimensional information fusion on the primary fault analysis results and the degradation trend conclusions to obtain the classification early warning information of the high-voltage power equipment, it is specifically used for: By jointly determining the results of the initial fault analysis and the conclusion of the deterioration trend, a fault confirmation conclusion is obtained for the high-voltage power equipment. Based on the traveling wave time difference signal in the synchronous heterogeneous dataset, the fault section of the cable line with abnormal fault in the high-voltage power equipment is defined to obtain the preliminary fault section range of the high-voltage power equipment. Based on the preliminary fault zone range, multi-source spatial information fusion is performed on the spatial center of the local thermal focusing phenomenon in the high-voltage power equipment, the specific location of the abnormal temperature node, and the location of the dominant signal source of the effective partial discharge event to obtain the physical coordinates of the fault point of the high-voltage power equipment. Based on the dominant abnormal signal characteristics in the primary fault analysis results, the nature of the abnormal fault is determined to obtain the fault type of the high-voltage power equipment. The fault confirmation conclusion, the physical coordinates of the fault point, and the fault type are encapsulated to obtain the classification and early warning information of the high-voltage power equipment.

8. The power high-voltage equipment monitoring and early warning system based on multi-parameter fusion analysis as described in claim 7, characterized in that, The formula for calculating the physical coordinates of the fault point is as follows: ； In the formula, The physical coordinates of the fault point are: The coordinates of the spatial center of the localized thermal focusing phenomenon are: These are the specific location coordinates of the abnormal temperature node. The coordinates are the location coordinates of the dominant signal source of the effective partial discharge event. Assign a preset confidence weight to the thermal focusing location. Assign a preset confidence weight to the temperature node location. The preset confidence weights for the location of the discharge signal source are used.

Images