Ultra-wideband PD intelligent detection system for AC and DC withstand voltage test

Abstract

The present invention discloses an ultra-wideband PD intelligent detection system for AC and DC withstand voltage tests, which relates to the field of measuring electrical variables, including: obtaining transformer insulation test cases, obtaining three-phase bushing ultra-wideband pulse signals, core grounding wire ultra-wideband pulse signals and neutral point grounding wire ultra-wideband pulse signals through withstand voltage tests. According to the transformer test model, the reference signal matching channel is matched, and each ultra-wideband reference pulse signal is output. By comparing each ultra-wideband pulse signal with the ultra-wideband reference pulse signal, the corresponding ultra-wideband pulse deviation matrix is ​​obtained. Through the local discharge anomaly detection channel, the matrix is ​​processed to obtain an insulation anomaly binary mark. When the insulation anomaly binary mark is 1, an insulation anomaly signal is generated. The defects of the transformer insulation state monitoring method in the prior art in terms of detection cycle, response speed and diagnostic accuracy are solved, and the technical problem of difficulty in realizing real-time and intelligent insulation anomaly detection is solved.

Description

Technical Field

[0001] The present application relates to the field of measuring electrical variables, and in particular to an ultra-wideband PD intelligent detection system for AC and DC withstand voltage tests. Background Art

[0002] With the continuous expansion of the scale of power systems and the increasingly complex operating conditions of power equipment, the health of the insulation system of transformers, as key equipment in power transmission and distribution, is directly related to the safety and stable operation of the entire power system. Insulation materials will gradually age due to factors such as voltage stress, ambient temperature, and humidity during long-term operation, resulting in insulation performance degradation and even partial discharge (PD), which are early signs of insulation failure. The pulse threshold signal of the existing PD detection process requires custom configuration by the client, which is time-consuming and costly, and stability cannot be guaranteed.

[0003] Therefore, the defects of the transformer insulation status monitoring method in the prior art in terms of detection cycle, response speed and diagnostic accuracy make it difficult to achieve the technical problem of real-time and intelligent insulation abnormality detection. Summary of the invention

[0004] This application solves the technical problem that the transformer insulation state monitoring method in the prior art has defects in detection cycle, response speed and diagnostic accuracy, and is difficult to achieve real-time and intelligent insulation abnormality detection by providing an ultra-wideband PD intelligent detection system for AC and DC withstand voltage tests. Through multi-sensor data acquisition, reference signal matching and Markov model analysis, highly accurate and real-time intelligent monitoring of insulation state is achieved, significantly improving the efficiency and reliability of insulation abnormality detection.

[0005] The present application provides an ultra-wideband PD intelligent detection system for AC and DC withstand voltage tests, wherein the system is communicatively connected with an ultra-wideband pulse current sensor, wherein the ultra-wideband pulse current sensor includes a three-phase bushing ultra-wideband pulse current sensor, an iron core grounding wire ultra-wideband pulse current sensor, and a neutral point grounding wire ultra-wideband pulse current sensor, and includes: a data extraction module, used to obtain transformer insulation test cases, wherein the transformer insulation test cases include transformer test model, transformer service time, and test insulation voltage; a pulse signal receiving module, used to communicate with the three-phase bushing ultra-wideband pulse current sensor, the iron core grounding wire ultra-wideband pulse current sensor, and the neutral point grounding wire ultra-wideband pulse current sensor and receive the three-phase bushing ultra-wideband pulse signal, the iron core grounding wire ultra-wideband pulse signal, and the neutral point grounding wire ultra-wideband pulse signal for withstand voltage tests under the test insulation voltage; a reference pulse signal acquisition module, used to match the reference signal matching channel according to the transformer test model, and to match the transformer service time, the test voltage, and the reference signal matching channel. The test insulation voltage is processed to output a three-phase bushing ultra-wideband reference pulse signal, an iron core grounding wire ultra-wideband reference pulse signal and a neutral point grounding wire ultra-wideband reference pulse signal; a deviation matrix acquisition module is used to compare the three-phase bushing ultra-wideband pulse signal with the three-phase bushing ultra-wideband reference pulse signal to generate a three-phase bushing ultra-wideband pulse deviation matrix, compare the iron core grounding wire ultra-wideband pulse signal with the iron core grounding wire ultra-wideband reference pulse signal to generate an iron core grounding wire ultra-wideband pulse deviation matrix, compare the neutral point grounding wire ultra-wideband pulse signal with the neutral point grounding wire ultra-wideband reference pulse signal to generate a neutral point grounding wire ultra-wideband pulse deviation matrix; a binary identification acquisition module is used to process the three-phase bushing ultra-wideband pulse deviation matrix, the iron core grounding wire ultra-wideband pulse deviation matrix and the neutral point grounding wire ultra-wideband pulse deviation matrix through a partial discharge abnormality detection channel to obtain an insulation abnormality binary identification; an abnormal signal acquisition module is used to generate an insulation abnormality signal when the insulation abnormality binary identification is 1.

[0006] In a possible implementation, the reference pulse signal acquisition module is further used to: process the test insulation voltage according to the initial reference signal matching layer to obtain an initial three-phase bushing ultra-wideband reference pulse signal, an initial iron core grounding wire ultra-wideband reference pulse signal and an initial neutral point grounding wire ultra-wideband reference pulse signal; when the service time of the transformer is greater than the service time threshold, according to the service attenuation analysis layer, the initial three-phase bushing ultra-wideband reference pulse signal, the initial iron core grounding wire ultra-wideband reference pulse signal and the initial neutral point grounding wire ultra-wideband reference pulse signal are matched based on the service time of the transformer. The service attenuation analysis is performed on the pulse signal to obtain the three-phase bushing ultra-wideband reference pulse signal, the core grounding wire ultra-wideband reference pulse signal and the neutral point grounding wire ultra-wideband reference pulse signal; when the service time of the transformer is less than or equal to the service time threshold, the initial three-phase bushing ultra-wideband reference pulse signal, the initial core grounding wire ultra-wideband reference pulse signal and the initial neutral point grounding wire ultra-wideband reference pulse signal are set as the three-phase bushing ultra-wideband reference pulse signal, the core grounding wire ultra-wideband reference pulse signal and the neutral point grounding wire ultra-wideband reference pulse signal.

[0007] In a possible implementation, the reference pulse signal acquisition module is further used to: based on the transformer test model as a constraint, collect a first voltage monitoring record data set with a service time less than or equal to the service time threshold, a first three-phase bushing ultra-wideband pulse signal record data set, a first core grounding wire ultra-wideband pulse signal record data set, and a first neutral point grounding wire ultra-wideband reference pulse signal record data set; based on the transformer test model and the first voltage monitoring record data set as constraints, collect a first service time record data set, a second three-phase bushing ultra-wideband pulse signal record data set, a second core grounding wire ultra-wideband pulse signal record data set, and a second neutral point grounding wire ultra-wideband reference pulse signal record data set with a service time greater than the service time threshold; based on the first three-phase bushing ultra-wideband pulse signal record data set, a second core grounding wire ultra-wideband pulse signal record data set, and a second neutral point grounding wire ultra-wideband reference pulse signal record data set The broadband pulse signal recording data set, the first iron core grounding wire ultra-wideband pulse signal recording data set and the first neutral point grounding wire ultra-wideband reference pulse signal recording data set are used as supervision, and the first voltage monitoring recording data set is used as input to train the initial reference signal matching layer; the second three-phase bushing ultra-wideband pulse signal recording data set, the second iron core grounding wire ultra-wideband pulse signal recording data set and the second neutral point grounding wire ultra-wideband reference pulse signal recording data set are used as supervision, and the first service time recording data set, the first three-phase bushing ultra-wideband pulse signal recording data set, the first iron core grounding wire ultra-wideband pulse signal recording data set and the first neutral point grounding wire ultra-wideband reference pulse signal recording data set are used as input to train the service attenuation analysis layer.

[0008] In a possible implementation, the reference pulse signal acquisition module is also used to: extract the first voltage monitoring record value according to the first voltage monitoring record data set; collect the first service time record value, the second three-phase bushing ultra-wideband pulse signal initial record value, the second core grounding wire ultra-wideband pulse signal initial record value and the second neutral point grounding wire ultra-wideband reference pulse signal initial record value based on the transformer test model and the first voltage monitoring record value as constraints; collect a number of second three-phase bushing ultra-wideband pulse signal incremental record values, a number of second core grounding wire ultra-wideband pulse signal incremental record values ​​and a number of second neutral point grounding wire ultra-wideband reference pulse signal incremental record values ​​based on the transformer test model, the first service time record value and the first voltage monitoring record value as constraints. The incremental recorded values ​​of the second three-phase bushing ultra-wideband pulse signal are evaluated together with the initial recorded values ​​of the second three-phase bushing ultra-wideband pulse signal to obtain the second three-phase bushing ultra-wideband pulse signal recorded data; the incremental recorded values ​​of the second iron core grounding wire ultra-wideband pulse signal are evaluated together with the initial recorded values ​​of the second iron core grounding wire ultra-wideband pulse signal to obtain the second iron core grounding wire ultra-wideband pulse signal recorded data; the incremental recorded values ​​of the second neutral point grounding wire ultra-wideband reference pulse signal are evaluated together with the initial recorded values ​​of the second neutral point grounding wire ultra-wideband reference pulse signal to obtain the second neutral point grounding wire ultra-wideband reference pulse signal recorded data.

[0009] In a possible implementation, the binary identification acquisition module is also used to: process the three-phase bushing ultra-wideband pulse deviation matrix through the three-phase bushing ultra-wideband pulse detection network, and output a first insulation abnormality binary identification; when the first insulation abnormality binary identification is 1, configure the insulation abnormality binary identification to 1; when the first insulation abnormality binary identification is 0, process the core grounding wire ultra-wideband pulse deviation matrix through the core grounding wire ultra-wideband pulse detection network, and output a second insulation abnormality binary identification; when the second insulation abnormality binary identification is 1, configure the insulation abnormality binary identification to 1; when the second insulation abnormality binary identification is 0, The point-to-ground wire ultra-wideband pulse detection network processes the neutral-point grounding wire ultra-wideband pulse deviation matrix and outputs a third insulation abnormality binary mark; when the third insulation abnormality binary mark is 1, the insulation abnormality binary mark is configured to 1; when the third insulation abnormality binary mark is 0, the three-phase bushing ultra-wideband pulse deviation matrix, the core grounding wire ultra-wideband pulse deviation matrix and the neutral-point grounding wire ultra-wideband pulse deviation matrix are processed through the ultra-wideband pulse global analysis network, and a fourth insulation abnormality binary mark is output; when the fourth insulation abnormality binary mark is 1, the insulation abnormality binary mark is configured to 1, otherwise, the insulation abnormality binary mark is configured to 0.

[0010] In a possible implementation, the binary identification acquisition module is also used to: configure a pulse signal deviation threshold; based on the pulse signal deviation threshold, constrain the ultra-wideband pulse recording deviation matrix of the core grounding wire, collect the first ultra-wideband pulse recording deviation matrix data set and the first insulation state detection identification data set; construct an insulation abnormality transfer probability matrix based on the first ultra-wideband pulse recording deviation matrix data set and the first insulation state detection identification data set; when the insulation abnormality transfer probability is greater than or equal to the abnormality transfer probability threshold, configure the insulation abnormality binary identification true value to be 1, otherwise, configure the insulation abnormality binary identification true value to be 0; construct a Markov chain based on the core grounding wire ultra-wideband pulse record deviation matrix data set, the insulation abnormality transfer probability matrix and the insulation abnormality binary identification true value; constrain the core grounding wire ultra-wideband pulse record deviation matrix based on the pulse signal deviation threshold, collect the second core grounding wire ultra-wideband pulse record deviation matrix data set and the second insulation state detection identification data set; take the second insulation state detection identification data set as supervision, take the second core grounding wire ultra-wideband pulse record deviation matrix data set as input, train the Markov model based on the Markov chain, and obtain the core grounding wire ultra-wideband pulse detection network.

[0011] In a possible implementation, the binary identification acquisition module is also used to: extract the first core grounding wire ultra-wideband pulse recording deviation matrix data according to the first core grounding wire ultra-wideband pulse recording deviation matrix data set; obtain a number of first insulation state detection identification data with the first core grounding wire ultra-wideband pulse recording deviation matrix data as a constraint; count the proportion of the number of the first insulation state detection identification data being 1, set it as the first insulation abnormality transfer probability, and add it to the insulation abnormality transfer probability matrix, which is a probability matrix that describes the insulation state changing from normal to abnormal.

[0012] In a possible implementation, the abnormal signal acquisition module is also used to: collect a three-phase bushing ultra-wideband pulse recording deviation matrix data set, an iron core grounding wire ultra-wideband pulse recording deviation matrix data set, a neutral point grounding wire ultra-wideband pulse recording deviation matrix data set and an insulation status detection identification data set, in which the first insulation abnormality binary identification, the second insulation abnormality binary identification and the third insulation abnormality binary identification are all 0, and train the ultra-wideband pulse global analysis network. The construction steps of the ultra-wideband pulse global analysis network and the iron core grounding wire ultra-wideband pulse detection network are the same.

[0013] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0014] The ultra-wideband PD intelligent detection system for AC and DC withstand voltage tests provided in the present application includes: obtaining a transformer insulation test case, and obtaining a three-phase bushing ultra-wideband pulse signal, an iron core grounding wire ultra-wideband pulse signal, and a neutral point grounding wire ultra-wideband pulse signal through a withstand voltage test. According to the transformer test model, the reference signal matching channel is matched, and each ultra-wideband reference pulse signal is output. By comparing each ultra-wideband pulse signal with the ultra-wideband reference pulse signal, the corresponding ultra-wideband pulse deviation matrix is ​​obtained. Through the local discharge anomaly detection channel, the matrix is ​​processed to obtain an insulation anomaly binary mark. When the insulation anomaly binary mark is 1, an insulation anomaly signal is generated. The defects of the transformer insulation state monitoring method in the prior art in terms of detection cycle, response speed and diagnostic accuracy, and the technical problem that it is difficult to realize real-time and intelligent insulation anomaly detection. Through multi-sensor data acquisition, reference signal matching and Markov model analysis, intelligent monitoring of the insulation state with high accuracy and real-time performance is achieved, which significantly improves the efficiency and reliability of insulation anomaly detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0016] Figure 1 A schematic diagram of the structure of an ultra-wideband PD intelligent detection system for AC and DC withstand voltage tests provided in an embodiment of the present application;

[0017] Figure 2 A schematic diagram of the process of obtaining the ultra-wideband reference pulse signal of the three-phase bushing, the ultra-wideband reference pulse signal of the core grounding wire and the ultra-wideband reference pulse signal of the neutral point grounding wire in the ultra-wideband PD intelligent detection system for AC and DC withstand voltage test of this application.

[0018] Explanation of the reference numerals: data extraction module 11 , pulse signal receiving module 12 , reference pulse signal acquisition module 13 , deviation matrix acquisition module 14 , binary identification acquisition module 15 , abnormal signal acquisition module 16 . DETAILED DESCRIPTION

[0019] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0020] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0021] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, systems, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0022] The present application embodiment provides an ultra-wideband PD intelligent detection system for AC and DC withstand voltage tests, such as Figure 1 As shown, the system comprises:

[0023] The data extraction module 11 is used to obtain transformer insulation test cases, wherein the transformer insulation test cases include transformer test model, transformer service time and test insulation voltage. The pulse signal receiving module 12 is used to communicate with the three-phase bushing ultra-wideband pulse current sensor, the core grounding wire ultra-wideband pulse current sensor and the neutral point grounding wire ultra-wideband pulse current sensor and receive the three-phase bushing ultra-wideband pulse signal, the core grounding wire ultra-wideband pulse signal and the neutral point grounding wire ultra-wideband pulse signal for withstand voltage test under the test insulation voltage. The reference pulse signal acquisition module 13 is used to match the reference signal matching channel according to the transformer test model, process the transformer service time and the test insulation voltage, and output the three-phase bushing ultra-wideband reference pulse signal, the core grounding wire ultra-wideband reference pulse signal and the neutral point grounding wire ultra-wideband reference pulse signal.

[0024] The data extraction module 11 is used to obtain transformer insulation test cases, wherein the transformer insulation test cases include transformer test model, transformer service time and test insulation voltage. The transformer test model indicates the specific model of the transformer, which is used to determine the test parameters and match the corresponding reference signal. The transformer service time refers to the cumulative working time of the transformer in actual operation, which affects the attenuation degree of its insulation performance. The test insulation voltage is the voltage value applied in the withstand voltage test, which is used to detect the performance of the insulating material under high voltage. Subsequently, the pulse signal receiving module 12 is used to communicate with the three-phase bushing ultra-wideband pulse current sensor, the core grounding wire ultra-wideband pulse current sensor and the neutral point grounding wire ultra-wideband pulse current sensor, and receive the three-phase bushing ultra-wideband pulse signal, the core grounding wire ultra-wideband pulse signal and the neutral point grounding wire ultra-wideband pulse signal of the withstand voltage test under the test insulation voltage. The three-phase bushing ultra-wideband pulse current sensor is an ultra-wideband pulse current sensor installed at the three-phase bushing of the transformer, which is used to monitor the partial discharge activity in the three-phase current. The ultra-wideband pulse current sensor for the core grounding wire is an ultra-wideband pulse current sensor installed at the core grounding wire, and is used to monitor the local discharge activity at the core grounding wire. The ultra-wideband pulse current sensor for the neutral point grounding wire is an ultra-wideband pulse current sensor installed at the neutral point grounding wire, and is used to monitor the local discharge activity at the neutral point grounding wire. The reference pulse signal acquisition module 13 is used to select and match the corresponding reference signal matching channel according to the transformer test model. The channel includes two main parts: the initial reference signal matching layer: processes the test insulation voltage and generates the initial reference pulse signal of the transformer in the initial state (i.e., the unserviced state). The service attenuation analysis layer performs attenuation analysis on the initial reference signal based on the service time, and outputs the reference pulse signal after attenuation processing. By processing the service time of the transformer and the test insulation voltage, the three-phase bushing ultra-wideband reference pulse signal, the core grounding wire ultra-wideband reference pulse signal and the neutral point grounding wire ultra-wideband reference pulse signal are output. The reference pulse signal is a reference reference signal corresponding to the current transformer state, so as to facilitate subsequent signal comparison and abnormality detection.

[0025] Further, such as Figure 2As shown, the reference pulse signal acquisition module 13 is also used to: process the test insulation voltage according to the initial reference signal matching layer to obtain the initial three-phase bushing ultra-wideband reference pulse signal, the initial core grounding wire ultra-wideband reference pulse signal and the initial neutral point grounding wire ultra-wideband reference pulse signal. When the service time of the transformer is greater than the service time threshold, according to the service attenuation analysis layer, the initial three-phase bushing ultra-wideband reference pulse signal, the initial core grounding wire ultra-wideband reference pulse signal and the initial neutral point grounding wire ultra-wideband reference pulse signal are subjected to service attenuation analysis based on the service time of the transformer to obtain the three-phase bushing ultra-wideband reference pulse signal, the core grounding wire ultra-wideband reference pulse signal and the neutral point grounding wire ultra-wideband reference pulse signal. When the service time of the transformer is less than or equal to the service time threshold, the initial three-phase bushing ultra-wideband reference pulse signal, the initial iron core grounding wire ultra-wideband reference pulse signal and the initial neutral point grounding wire ultra-wideband reference pulse signal are set as the three-phase bushing ultra-wideband reference pulse signal, the iron core grounding wire ultra-wideband reference pulse signal and the neutral point grounding wire ultra-wideband reference pulse signal.

[0026] Outputting three-phase bushing ultra-wideband reference pulse signals, core grounding wire ultra-wideband reference pulse signals and neutral point grounding wire ultra-wideband reference pulse signals, including: processing the test insulation voltage according to the initial reference signal matching layer, obtaining the initial three-phase bushing ultra-wideband reference pulse signal, initial core grounding wire ultra-wideband reference pulse signal and initial neutral point grounding wire ultra-wideband reference pulse signal corresponding to the transformer test model. The reference pulse signal is an initial reference pulse signal generated by the test insulation voltage without considering the influence of the service time, which serves as the basis for subsequent analysis. When the service time of the transformer exceeds the preset service time threshold, the system needs to perform attenuation analysis on the initial reference pulse signal through the service attenuation analysis layer to reflect the insulation performance attenuation caused by long-term operation. The preset service time threshold is a preset service time limit, which is used to determine whether attenuation analysis is required. If the threshold is exceeded, attenuation processing is required, and if it is not exceeded, no attenuation processing is required. When performing attenuation processing, according to the service attenuation analysis layer, based on the service time of the transformer, the initial three-phase bushing ultra-wideband reference pulse signal, the initial core grounding wire ultra-wideband reference pulse signal and the initial neutral point grounding wire ultra-wideband reference pulse signal are subjected to service attenuation analysis to obtain the three-phase bushing ultra-wideband reference pulse signal, the core grounding wire ultra-wideband reference pulse signal and the neutral point grounding wire ultra-wideband reference pulse signal. When the service time of the transformer is less than or equal to the service time threshold, the attenuation effect of service is relatively weak, and no attenuation analysis is required. The initial three-phase bushing ultra-wideband reference pulse signal, the initial core grounding wire ultra-wideband reference pulse signal and the initial neutral point grounding wire ultra-wideband reference pulse signal are set as the three-phase bushing ultra-wideband reference pulse signal, the core grounding wire ultra-wideband reference pulse signal and the neutral point grounding wire ultra-wideband reference pulse signal.

[0027] Further, the reference pulse signal acquisition module 13 is also used to: with the transformer test model as a constraint, collect the first voltage monitoring record data set, the first three-phase bushing ultra-wideband pulse signal record data set, the first core grounding wire ultra-wideband pulse signal record data set, and the first neutral point grounding wire ultra-wideband reference pulse signal record data set, whose service time is less than or equal to the service time threshold. With the transformer test model and the first voltage monitoring record data set as constraints, collect the first service time record data set, the second three-phase bushing ultra-wideband pulse signal record data set, the second core grounding wire ultra-wideband pulse signal record data set, and the second neutral point grounding wire ultra-wideband reference pulse signal record data set, whose service time is greater than the service time threshold. With the first three-phase bushing ultra-wideband pulse signal record data set, the first core grounding wire ultra-wideband pulse signal record data set, and the first neutral point grounding wire ultra-wideband reference pulse signal record data set as supervision, and the first voltage monitoring record data set as input, train the initial reference signal matching layer. The service attenuation analysis layer is trained with the second three-phase bushing ultra-wideband pulse signal recording data set, the second iron core grounding wire ultra-wideband pulse signal recording data set and the second neutral point grounding wire ultra-wideband reference pulse signal recording data set as supervision, and the first service time recording data set, the first three-phase bushing ultra-wideband pulse signal recording data set, the first iron core grounding wire ultra-wideband pulse signal recording data set and the first neutral point grounding wire ultra-wideband reference pulse signal recording data set as input.

[0028] The reference signal matching channel construction step includes: taking the transformer test model as a constraint, collecting the first voltage monitoring record data set, i.e., the voltage monitoring data recorded under the test insulation voltage, the first three-phase bushing ultra-wideband pulse signal record data set, the first core grounding wire ultra-wideband pulse signal record data set, and the first neutral point grounding wire ultra-wideband reference pulse signal record data set, whose service time is less than or equal to the service time threshold, and taking the transformer test model and the first voltage monitoring record data set as constraints, collecting the first service time record data set, the second three-phase bushing ultra-wideband pulse signal record data set, the second core grounding wire ultra-wideband pulse signal record data set, and the second neutral point grounding wire ultra-wideband reference pulse signal record data set, whose service time is greater than the service time threshold.

[0029] The first three-phase bushing ultra-wideband pulse signal recording data set, the first core grounding wire ultra-wideband pulse signal recording data set and the first neutral point grounding wire ultra-wideband reference pulse signal recording data set are used as supervision, and the first voltage monitoring recording data set is used as input to train the initial reference signal matching layer. The initial reference signal matching layer is constructed based on a neural network model, and the first voltage monitoring recording data set is input into an untrained neural network model. The supervised training is performed through the supervision data until the model output accuracy meets the preset accuracy rate. The model training is completed to obtain the initial reference signal matching layer of the corresponding transformer test model. Further, the second three-phase bushing ultra-wideband pulse signal recording data set, the second core grounding wire ultra-wideband pulse signal recording data set and the second neutral point grounding wire ultra-wideband reference pulse signal recording data set are used as supervision, and the first service time recording data set, the first three-phase bushing ultra-wideband pulse signal recording data set, the first core grounding wire ultra-wideband pulse signal recording data set and the first neutral point grounding wire ultra-wideband reference pulse signal recording data set are used as input, and the service attenuation analysis layer is trained in the same training method as the initial reference signal matching layer.

[0030] Furthermore, the reference pulse signal acquisition module 13 is also used to: extract the first voltage monitoring record value according to the first voltage monitoring record data set. Taking the transformer test model and the first voltage monitoring record value as constraints, collect the first service time record value, the second three-phase bushing ultra-wideband pulse signal initial record value, the second core grounding wire ultra-wideband pulse signal initial record value, and the second neutral point grounding wire ultra-wideband reference pulse signal initial record value. Taking the transformer test model, the first service time record value and the first voltage monitoring record value as constraints, collect several second three-phase bushing ultra-wideband pulse signal incremental record values, several second core grounding wire ultra-wideband pulse signal incremental record values, and several second neutral point grounding wire ultra-wideband reference pulse signal incremental record values. A centralized value evaluation is performed on the several second three-phase bushing ultra-wideband pulse signal incremental record values ​​and the second three-phase bushing ultra-wideband pulse signal initial record values ​​to obtain the second three-phase bushing ultra-wideband pulse signal record data; a centralized value evaluation is performed on the several second iron core grounding wire ultra-wideband pulse signal incremental record values ​​and the second iron core grounding wire ultra-wideband pulse signal initial record values ​​to obtain the second iron core grounding wire ultra-wideband pulse signal record data; a centralized value evaluation is performed on the several second neutral point grounding wire ultra-wideband reference pulse signal incremental record values ​​and the second neutral point grounding wire ultra-wideband reference pulse signal initial record values ​​to obtain the second neutral point grounding wire ultra-wideband reference pulse signal record data.

[0031] Taking the transformer test model and the first voltage monitoring record data set as constraints, collecting the first service time record data set, the second three-phase bushing ultra-wideband pulse signal record data set, the second core grounding wire ultra-wideband pulse signal record data set, and the second neutral point grounding wire ultra-wideband reference pulse signal record data set, which have a service time greater than the service time threshold, including: extracting the first voltage monitoring record value according to the first voltage monitoring record data set, and collecting the first service time record value, the second three-phase bushing ultra-wideband pulse signal initial record value, the second core grounding wire ultra-wideband pulse signal initial record value, and the second neutral point grounding wire ultra-wideband reference pulse signal initial record value based on the transformer operation test big data and taking the transformer test model and the first voltage monitoring record value as constraints. The second three-phase bushing ultra-wideband pulse signal initial record value, the second core grounding wire ultra-wideband pulse signal initial record value, and the second neutral point grounding wire ultra-wideband reference pulse signal initial record value correspond to the first service time record value. The initial record value is the reference ultra-wideband pulse signal initial record value of the transformer under the corresponding voltage monitoring record value condition of the transformer test model, and the initial record value is the standard parameter obtained under the test condition without being affected by the actual operating environment. Since the transformer will be affected by various external factors such as the environment during actual operation, there will be certain fluctuations in the initial recorded value. By taking the transformer test model, the first service time recorded value and the first voltage monitoring recorded value as constraints, several second three-phase bushing ultra-wideband pulse signal incremental recorded values, several second core grounding wire ultra-wideband pulse signal incremental recorded values ​​and several second neutral point grounding wire ultra-wideband reference pulse signal incremental recorded values ​​are collected. The incremental recorded values ​​are deviation data from the initial recorded values ​​obtained by the transformer under different operating environments. The plurality of second three-phase bushing ultra-wideband pulse signal incremental record values ​​and the second three-phase bushing ultra-wideband pulse signal initial record values ​​are evaluated for centralized values ​​to obtain the second three-phase bushing ultra-wideband pulse signal record data, the plurality of second core grounding wire ultra-wideband pulse signal incremental record values ​​and the second core grounding wire ultra-wideband pulse signal initial record values ​​are evaluated for centralized values ​​to obtain the second core grounding wire ultra-wideband pulse signal record data, the plurality of second neutral point grounding wire ultra-wideband reference pulse signal incremental record values ​​and the second neutral point grounding wire ultra-wideband reference pulse signal initial record values ​​are evaluated for centralized values ​​to obtain the second neutral point grounding wire ultra-wideband reference pulse signal record data. The centralized value evaluation is to obtain the mode of the data in the second three-phase bushing ultra-wideband pulse signal incremental record value and the second three-phase bushing ultra-wideband pulse signal initial record value, and to add the mode to generate the corresponding pulse signal record data. The corresponding pulse signal record data under different service time record values ​​are obtained to obtain each pulse signal record data set.

[0032] The deviation matrix acquisition module 14 is used to compare the three-phase bushing ultra-wideband pulse signal with the three-phase bushing ultra-wideband reference pulse signal to generate the three-phase bushing ultra-wideband pulse deviation matrix, compare the core grounding wire ultra-wideband pulse signal with the core grounding wire ultra-wideband reference pulse signal to generate the core grounding wire ultra-wideband pulse deviation matrix, and compare the neutral point grounding wire ultra-wideband pulse signal with the neutral point grounding wire ultra-wideband reference pulse signal to generate the neutral point grounding wire ultra-wideband pulse deviation matrix. The binary identification acquisition module 15 is used to process the three-phase bushing ultra-wideband pulse deviation matrix, the core grounding wire ultra-wideband pulse deviation matrix and the neutral point grounding wire ultra-wideband pulse deviation matrix through the partial discharge abnormality detection channel to obtain the insulation abnormality binary identification. The abnormal signal acquisition module 16 is used to generate an insulation abnormality signal when the insulation abnormality binary identification is 1.

[0033] The deviation matrix acquisition module 14 is used to compare the three-phase bushing ultra-wideband pulse signal with the three-phase bushing ultra-wideband reference pulse signal to generate the three-phase bushing ultra-wideband pulse deviation matrix, compare the core grounding wire ultra-wideband pulse signal with the core grounding wire ultra-wideband reference pulse signal to generate the core grounding wire ultra-wideband pulse deviation matrix, and compare the neutral point grounding wire ultra-wideband pulse signal with the neutral point grounding wire ultra-wideband reference pulse signal to generate the neutral point grounding wire ultra-wideband pulse deviation matrix. The ultra-wideband pulse deviation matrix is ​​a matrix composed of the deviations generated by the pulse signal. The binary identification acquisition module 15 is used to process each matrix in turn through the local discharge abnormality detection channel to obtain the insulation abnormality binary identification. The abnormal signal acquisition module 16 is used to generate an insulation abnormality signal when the insulation abnormality binary identification is 1.

[0034] Furthermore, the binary identification acquisition module 15 is also used to: process the three-phase bushing ultra-wideband pulse deviation matrix through the three-phase bushing ultra-wideband pulse detection network, and output a first insulation abnormality binary identification. When the first insulation abnormality binary identification is 1, the insulation abnormality binary identification is configured to 1. When the first insulation abnormality binary identification is 0, the core grounding wire ultra-wideband pulse deviation matrix is ​​processed through the core grounding wire ultra-wideband pulse detection network, and a second insulation abnormality binary identification is output. When the second insulation abnormality binary identification is 1, the insulation abnormality binary identification is configured to 1. When the second insulation abnormality binary identification is 0, the neutral point grounding wire ultra-wideband pulse deviation matrix is ​​processed through the neutral point grounding wire ultra-wideband pulse detection network, and a third insulation abnormality binary identification is output. When the third insulation abnormality binary identification is 1, the insulation abnormality binary identification is configured to 1. When the third insulation abnormality binary identification is 0, the three-phase bushing ultra-wideband pulse deviation matrix, the core grounding wire ultra-wideband pulse deviation matrix and the neutral point grounding wire ultra-wideband pulse deviation matrix are processed through the ultra-wideband pulse global analysis network to output the fourth insulation abnormality binary identification. When the fourth insulation abnormality binary identification is 1, the insulation abnormality binary identification is configured to 1, otherwise, the insulation abnormality binary identification is configured to 0. The defects of the transformer insulation state monitoring method in the prior art in terms of detection cycle, response speed and diagnostic accuracy are solved, and the technical problem of difficulty in realizing real-time and intelligent insulation abnormality detection is solved. Through multi-sensor data acquisition, reference signal matching and Markov model analysis, intelligent monitoring of the insulation state with high accuracy and real-time performance is achieved, which significantly improves the efficiency and reliability of insulation abnormality detection.

[0035] Among them, the steps of constructing the ultra-wideband pulse global analysis network include: collecting a three-phase bushing ultra-wideband pulse recording deviation matrix data set, an iron core grounding wire ultra-wideband pulse recording deviation matrix data set, a neutral point grounding wire ultra-wideband pulse recording deviation matrix data set and an insulation status detection identification data set, in which the first insulation abnormality binary identification, the second insulation abnormality binary identification and the third insulation abnormality binary identification are all 0, and training the ultra-wideband pulse global analysis network. The construction steps of the ultra-wideband pulse global analysis network are the same as those of the iron core grounding wire ultra-wideband pulse detection network.

[0036] Obtaining the insulation abnormality binary identification includes: processing the three-phase bushing ultra-wideband pulse deviation matrix through the three-phase bushing ultra-wideband pulse detection network, and outputting a first insulation abnormality binary identification. When the first insulation abnormality binary identification is 1, the insulation abnormality binary identification is configured to 1, and insulation abnormality exists at this time. When the first insulation abnormality binary identification is 0, there is no insulation abnormality at this time. Then continue to process the iron core grounding wire ultra-wideband pulse deviation matrix through the iron core grounding wire ultra-wideband pulse detection network, and output a second insulation abnormality binary identification. When the second insulation abnormality binary identification is 1, the insulation abnormality binary identification is configured to. When the second insulation abnormality binary identification is 0, the neutral point grounding wire ultra-wideband pulse deviation matrix is ​​processed through the neutral point grounding wire ultra-wideband pulse detection network, and a third insulation abnormality binary identification is output. When the third insulation abnormality binary identification is 1, the insulation abnormality binary identification is configured to 1. By detecting the insulation abnormalities in sequence, targeted abnormalities can be quickly discovered, thereby improving the efficiency of abnormality discovery. When no abnormality is found after sequential detection, the corresponding abnormality may be a combined abnormality. Then, the ultra-wideband pulse global analysis network is used to process the ultra-wideband pulse deviation matrix of the three-phase bushing, the ultra-wideband pulse deviation matrix of the core grounding wire, and the ultra-wideband pulse deviation matrix of the neutral point grounding wire, and output the fourth insulation abnormality binary mark. When the fourth insulation abnormality binary mark is 1, the insulation abnormality binary mark is configured to 1, otherwise, the insulation abnormality binary mark is configured to 0. When the insulation abnormality binary mark is configured to 0, the ultra-wideband pulse global analysis network is used to process the ultra-wideband pulse deviation matrix of the three-phase bushing, the ultra-wideband pulse deviation matrix of the core grounding wire, and the ultra-wideband pulse deviation matrix of the neutral point grounding wire, and output the insulation abnormality binary mark.

[0037] Furthermore, the binary identification acquisition module 15 is also used to: configure a pulse signal deviation threshold. Based on the pulse signal deviation threshold, the ultra-wideband pulse recording deviation matrix of the core grounding wire is constrained, and the first ultra-wideband pulse recording deviation matrix data set and the first insulation state detection identification data set are collected. According to the first ultra-wideband pulse recording deviation matrix data set of the core grounding wire and the first insulation state detection identification data set, an insulation abnormality transfer probability matrix is ​​constructed. When the insulation abnormality transfer probability is greater than or equal to the abnormality transfer probability threshold, the insulation abnormality binary identification truth value is configured to be 1, otherwise, the insulation abnormality binary identification truth value is configured to be 0. Based on the ultra-wideband pulse recording deviation matrix data set of the core grounding wire, the insulation abnormality transfer probability matrix and the insulation abnormality binary identification truth value, a Markov chain is constructed. Based on the pulse signal deviation threshold, the ultra-wideband pulse recording deviation matrix of the core grounding wire is constrained, and the second ultra-wideband pulse recording deviation matrix data set and the second insulation state detection identification data set of the core grounding wire are collected. The second insulation state detection identification data set is used as supervision, the second core grounding wire ultra-wideband pulse record deviation matrix data set is used as input, and a Markov model is trained based on the Markov chain to obtain the core grounding wire ultra-wideband pulse detection network.

[0038] The construction methods of the three-phase bushing ultra-wideband pulse detection network, the core grounding wire ultra-wideband pulse detection network, the neutral point grounding wire ultra-wideband pulse detection network and the ultra-wideband pulse global analysis network are the same, and only the training data are different. The construction of the neutral point grounding wire ultra-wideband pulse detection network is based on the neutral point grounding wire ultra-wideband pulse record deviation matrix, and the construction of the three-phase bushing ultra-wideband pulse detection network is based on the three-phase bushing ultra-wideband pulse detection record deviation matrix. Taking the construction of the core grounding wire ultra-wideband pulse detection network as an example, by configuring the pulse signal deviation threshold, the threshold is used to distinguish the normal state from the existing insulation abnormal state. The threshold is determined based on historical data and expert experience to ensure that abnormal discharge signals can be effectively identified in actual detection. Based on the pulse signal deviation threshold, the core grounding wire ultra-wideband pulse record deviation matrix is ​​constrained, and the first core grounding wire ultra-wideband pulse record deviation matrix data set and the first insulation state detection identification data set are collected. The first core grounding wire ultra-wideband pulse record deviation matrix data set is the data of the core grounding wire ultra-wideband pulse record deviation matrix that meets the constraints. The first insulation state detection identification data set is the abnormal or normal identification of the corresponding data insulation state.

[0039] By extracting all deviation values ​​from the first core grounding wire ultra-wideband pulse recording deviation matrix data set, calculating the proportion of 1 (abnormal) in the insulation state detection identification data set as the insulation abnormality transfer probability, the calculated insulation abnormality transfer probability is filled into the transfer probability matrix to construct the insulation abnormality transfer probability matrix. The insulation abnormality transfer probability matrix is ​​a probability matrix that describes the insulation state from normal to abnormal, and is used to predict future insulation state changes. When the insulation abnormality transfer probability is greater than or equal to the abnormality transfer probability threshold, the insulation abnormality binary identification true value is configured to be 1, otherwise, the insulation abnormality binary identification true value is configured to be 0. Based on the first core grounding wire ultra-wideband pulse recording deviation matrix data set, the insulation abnormality transfer probability matrix and the insulation abnormality binary identification true value, a Markov chain model is constructed to describe the dynamic change of the insulation state. Further, the data is re-collected and trained based on the above-mentioned evaluation criteria to obtain the final convergence model. Based on the pulse signal deviation threshold, the core grounding wire ultra-wideband pulse recording deviation matrix is ​​constrained, and the second core grounding wire ultra-wideband pulse recording deviation matrix data set and the second insulation state detection identification data set are collected. The second insulation state detection identification data set is supervised, and the second core grounding wire ultra-wideband pulse record deviation matrix data set is input. Based on the Markov chain, a Markov model is trained to obtain the core grounding wire ultra-wideband pulse detection network.

[0040] Furthermore, the binary identification acquisition module 15 is also used to: extract the first core grounding wire ultra-wideband pulse recording deviation matrix data according to the first core grounding wire ultra-wideband pulse recording deviation matrix data set. Take the first core grounding wire ultra-wideband pulse recording deviation matrix data as a constraint to obtain a number of first insulation state detection identification data; count the proportion of the number of the first insulation state detection identification data being 1, set it as the first insulation abnormality transfer probability, and add it into the insulation abnormality transfer probability matrix.

[0041] Constructing an insulation abnormality transfer probability matrix, including: extracting all relevant pulse deviation data according to the first core grounding wire ultra-wideband pulse recording deviation matrix data set to obtain the first core grounding wire ultra-wideband pulse recording deviation matrix data. And obtaining a number of corresponding first insulation state detection identification data. Counting the proportion of the number of the first insulation state detection identification data being 1, setting it as the first insulation abnormality transfer probability, and adding it to the insulation abnormality transfer probability matrix.

[0042] The embodiment of the present application obtains transformer insulation test cases through a data extraction module, and obtains ultra-wideband pulse signals of three-phase bushings, ultra-wideband pulse signals of core grounding wires, and ultra-wideband pulse signals of neutral point grounding wires through withstand voltage tests. According to the transformer test model, the reference signal matching channel is matched, and each ultra-wideband reference pulse signal is output. The corresponding ultra-wideband pulse deviation matrix is ​​obtained by comparing each ultra-wideband pulse signal with the ultra-wideband reference pulse signal. The matrix is ​​processed through a local discharge anomaly detection channel to obtain an insulation anomaly binary mark. When the insulation anomaly binary mark is 1, an insulation anomaly signal is generated. The defects of the transformer insulation state monitoring method in the prior art in terms of detection cycle, response speed, and diagnostic accuracy are solved, and the technical problem that it is difficult to realize real-time and intelligent insulation anomaly detection. Through multi-sensor data acquisition, reference signal matching, and Markov model analysis, intelligent monitoring of the insulation state with high accuracy and real-time performance is achieved, which significantly improves the efficiency and reliability of insulation anomaly detection.

[0043] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. An ultra-wideband PD intelligent detection system for AC and DC withstand voltage tests, the system being communicatively connected to an ultra-wideband pulse current sensor, the ultra-wideband pulse current sensor comprising a three-phase bushing ultra-wideband pulse current sensor, an iron core grounding wire ultra-wideband pulse current sensor, and a neutral point grounding wire ultra-wideband pulse current sensor, comprising: A data extraction module is used to obtain transformer insulation test cases, wherein the transformer insulation test cases include transformer test model, transformer service time and test insulation voltage; A pulse signal receiving module, used for communicating with the three-phase bushing ultra-wideband pulse current sensor, the core grounding wire ultra-wideband pulse current sensor and the neutral point grounding wire ultra-wideband pulse current sensor and receiving the three-phase bushing ultra-wideband pulse signal, the core grounding wire ultra-wideband pulse signal and the neutral point grounding wire ultra-wideband pulse signal subjected to the withstand voltage test under the test insulation voltage; A reference pulse signal acquisition module is used to match the reference signal matching channel according to the transformer test model, process the transformer service life and the test insulation voltage, and output a three-phase bushing ultra-wideband reference pulse signal, an iron core grounding wire ultra-wideband reference pulse signal, and a neutral point grounding wire ultra-wideband reference pulse signal; A deviation matrix acquisition module is used to compare the three-phase bushing ultra-wideband pulse signal with the three-phase bushing ultra-wideband reference pulse signal to generate a three-phase bushing ultra-wideband pulse deviation matrix, compare the core grounding wire ultra-wideband pulse signal with the core grounding wire ultra-wideband reference pulse signal to generate a core grounding wire ultra-wideband pulse deviation matrix, compare the neutral point grounding wire ultra-wideband pulse signal with the neutral point grounding wire ultra-wideband reference pulse signal to generate a neutral point grounding wire ultra-wideband pulse deviation matrix; A binary identification acquisition module is used to process the three-phase bushing ultra-wideband pulse deviation matrix, the core grounding wire ultra-wideband pulse deviation matrix and the neutral point grounding wire ultra-wideband pulse deviation matrix through a partial discharge abnormality detection channel to obtain an insulation abnormality binary identification; The abnormal signal acquisition module is used to generate an insulation abnormal signal when the insulation abnormal binary mark is 1.

2. The system according to claim 1, characterized in that The reference pulse signal acquisition module is also used for: According to the initial reference signal matching layer, the test insulation voltage is processed to obtain an initial three-phase bushing ultra-wideband reference pulse signal, an initial core grounding wire ultra-wideband reference pulse signal and an initial neutral point grounding wire ultra-wideband reference pulse signal; When the service time of the transformer is greater than the service time threshold, according to the service attenuation analysis layer, based on the service time of the transformer, the initial three-phase bushing ultra-wideband reference pulse signal, the initial core grounding wire ultra-wideband reference pulse signal and the initial neutral point grounding wire ultra-wideband reference pulse signal are subjected to service attenuation analysis to obtain the three-phase bushing ultra-wideband reference pulse signal, the core grounding wire ultra-wideband reference pulse signal and the neutral point grounding wire ultra-wideband reference pulse signal; When the service time of the transformer is less than or equal to the service time threshold, the initial three-phase bushing ultra-wideband reference pulse signal, the initial iron core grounding wire ultra-wideband reference pulse signal and the initial neutral point grounding wire ultra-wideband reference pulse signal are set as the three-phase bushing ultra-wideband reference pulse signal, the iron core grounding wire ultra-wideband reference pulse signal and the neutral point grounding wire ultra-wideband reference pulse signal.

3. The system according to claim 2, characterized in that The reference pulse signal acquisition module is also used for: Taking the transformer test model as a constraint, collecting a first voltage monitoring record data set whose service time is less than or equal to the service time threshold, a first three-phase bushing ultra-wideband pulse signal record data set, a first core grounding wire ultra-wideband pulse signal record data set, and a first neutral point grounding wire ultra-wideband reference pulse signal record data set; Taking the transformer test model and the first voltage monitoring record data set as constraints, collecting a first service time record data set whose service time is greater than the service time threshold, a second three-phase bushing ultra-wideband pulse signal record data set, a second core grounding wire ultra-wideband pulse signal record data set, and a second neutral point grounding wire ultra-wideband reference pulse signal record data set; The initial reference signal matching layer is trained by taking the first three-phase bushing ultra-wideband pulse signal record data set, the first core grounding wire ultra-wideband pulse signal record data set and the first neutral point grounding wire ultra-wideband reference pulse signal record data set as supervision and taking the first voltage monitoring record data set as input; The service attenuation analysis layer is trained with the second three-phase bushing ultra-wideband pulse signal recording data set, the second iron core grounding wire ultra-wideband pulse signal recording data set and the second neutral point grounding wire ultra-wideband reference pulse signal recording data set as supervision, and the first service time recording data set, the first three-phase bushing ultra-wideband pulse signal recording data set, the first iron core grounding wire ultra-wideband pulse signal recording data set and the first neutral point grounding wire ultra-wideband reference pulse signal recording data set as input.

4. The system according to claim 3, characterized in that The reference pulse signal acquisition module is also used for: Extracting a first voltage monitoring record value according to the first voltage monitoring record data set; Taking the transformer test model and the first voltage monitoring record value as constraints, collect the first service time record value, the second three-phase bushing ultra-wideband pulse signal initial record value, the second core grounding wire ultra-wideband pulse signal initial record value, and the second neutral point grounding wire ultra-wideband reference pulse signal initial record value; Taking the transformer test model, the first service time record value and the first voltage monitoring record value as constraints, collect a number of second three-phase bushing ultra-wideband pulse signal incremental record values, a number of second core grounding wire ultra-wideband pulse signal incremental record values ​​and a number of second neutral point grounding wire ultra-wideband reference pulse signal incremental record values; A centralized value evaluation is performed on the several second three-phase bushing ultra-wideband pulse signal incremental record values ​​and the second three-phase bushing ultra-wideband pulse signal initial record values ​​to obtain the second three-phase bushing ultra-wideband pulse signal record data; a centralized value evaluation is performed on the several second iron core grounding wire ultra-wideband pulse signal incremental record values ​​and the second iron core grounding wire ultra-wideband pulse signal initial record values ​​to obtain the second iron core grounding wire ultra-wideband pulse signal record data; a centralized value evaluation is performed on the several second neutral point grounding wire ultra-wideband reference pulse signal incremental record values ​​and the second neutral point grounding wire ultra-wideband reference pulse signal initial record values ​​to obtain the second neutral point grounding wire ultra-wideband reference pulse signal record data.

5. The system according to claim 1, wherein: The binary identification acquisition module is also used for: The three-phase bushing ultra-wideband pulse detection network is used to process the three-phase bushing ultra-wideband pulse deviation matrix, and a first insulation abnormality binary mark is output; When the first insulation abnormality binary flag is 1, configuring the insulation abnormality binary flag to be 1; When the first insulation abnormality binary mark is 0, the core grounding wire ultra-wideband pulse deviation matrix is ​​processed through the core grounding wire ultra-wideband pulse detection network to output a second insulation abnormality binary mark; When the second insulation abnormality binary flag is 1, the insulation abnormality binary flag is configured as 1; When the second insulation abnormality binary mark is 0, the neutral point grounding wire ultra-wideband pulse deviation matrix is ​​processed through the neutral point grounding wire ultra-wideband pulse detection network to output a third insulation abnormality binary mark; When the third insulation abnormality binary flag is 1, the insulation abnormality binary flag is configured to be 1; When the third insulation abnormality binary mark is 0, the three-phase bushing ultra-wideband pulse deviation matrix, the core grounding wire ultra-wideband pulse deviation matrix and the neutral point grounding wire ultra-wideband pulse deviation matrix are processed through the ultra-wideband pulse global analysis network to output a fourth insulation abnormality binary mark; When the fourth insulation abnormality binary flag is 1, the insulation abnormality binary flag is configured as 1; otherwise, the insulation abnormality binary flag is configured as 0.

6. The system according to claim 5, characterized in that The binary identification acquisition module is also used for: Configure the pulse signal deviation threshold; Based on the pulse signal deviation threshold, the core grounding wire ultra-wideband pulse recording deviation matrix is ​​constrained, and a first core grounding wire ultra-wideband pulse recording deviation matrix data set and a first insulation state detection identification data set are collected; Constructing an insulation abnormality transfer probability matrix according to the first core grounding wire ultra-wideband pulse recording deviation matrix data set and the first insulation state detection identification data set; When the insulation abnormality transfer probability is greater than or equal to the abnormality transfer probability threshold, the insulation abnormality binary identification true value is configured to be 1, otherwise, the insulation abnormality binary identification true value is configured to be 0; Constructing a Markov chain based on the core grounding wire ultra-wideband pulse recording deviation matrix data set, the insulation anomaly transfer probability matrix and the insulation anomaly binary identification true value; Based on the pulse signal deviation threshold, the core grounding wire ultra-wideband pulse recording deviation matrix is ​​constrained, and a second core grounding wire ultra-wideband pulse recording deviation matrix data set and a second insulation state detection identification data set are collected; The second insulation state detection identification data set is used as supervision, the second core grounding wire ultra-wideband pulse record deviation matrix data set is used as input, and a Markov model is trained based on the Markov chain to obtain the core grounding wire ultra-wideband pulse detection network.

7. The system according to claim 6, characterized in that The binary identification acquisition module is also used for: Extracting first core grounding wire ultra-wideband pulse recording deviation matrix data according to the first core grounding wire ultra-wideband pulse recording deviation matrix data set; Taking the first core grounding wire ultra-wideband pulse recording deviation matrix data as a constraint, obtaining a plurality of first insulation state detection identification data; The proportion of the first insulation state detection identification data being 1 is counted, set as the first insulation abnormality transfer probability, and added into the insulation abnormality transfer probability matrix, which is a probability matrix describing the insulation state changing from normal to abnormal.

8. The system according to claim 6, characterized in that The abnormal signal acquisition module is also used for: A three-phase bushing ultra-wideband pulse recording deviation matrix data set, an iron core grounding wire ultra-wideband pulse recording deviation matrix data set, a neutral point grounding wire ultra-wideband pulse recording deviation matrix data set and an insulation status detection identification data set, in which the first insulation abnormality binary identification, the second insulation abnormality binary identification and the third insulation abnormality binary identification are all 0, are collected to train the ultra-wideband pulse global analysis network. The construction steps of the ultra-wideband pulse global analysis network and the iron core grounding wire ultra-wideband pulse detection network are the same.

Images