A method and system for online monitoring of partial discharge of high-voltage switch cabinet

Abstract

The present application provides a method and system for online monitoring of partial discharge of high-voltage switch cabinets, which relates to the field of partial discharge monitoring technology. The method includes: performing structural division processing on the high-voltage switch cabinet to obtain a set of structural parts; calling a predetermined key part decision strategy to determine the key part set; dynamically monitoring the target partial discharge sensor deployed at the target part to obtain a target partial discharge signal; inputting the target partial discharge feature of the target partial discharge signal into a partial discharge classifier to obtain a target partial discharge judgment result; according to the target partial discharge judgment result, if the target part meets the predetermined conditions, issuing an early warning instruction; and performing a partial discharge early warning on the target part of the high-voltage switch cabinet according to the early warning instruction. This application can solve the problems of long cycle, large workload, and poor real-time performance of offline monitoring methods, realize real-time monitoring of partial discharge of high-voltage switch cabinets, and improve the real-time performance of the monitoring system.

Description

Technical Field

[0001] The present application relates to the technical field of partial discharge monitoring, and in particular to an online partial discharge monitoring method and system for a high-voltage switch cabinet. Background Art

[0002] High-voltage switchgear is an indispensable key equipment in the power system. Its main function is to realize the functions of circuit connection, disconnection, load switching, etc. in the power system. With the rapid development of social economy, the load demand of the power system continues to grow, and higher requirements are placed on the reliability, safety and stability of high-voltage switchgear. However, in the long-term operation process, partial discharge is prone to occur inside the high-voltage switchgear, which will cause the insulation performance of the equipment to deteriorate and even cause serious accidents. At present, the monitoring method for partial discharge of high-voltage switchgear is mainly offline monitoring, that is, through regular power outage inspection and detection instruments to detect equipment and find partial discharge defects. However, this method has problems such as long cycle, large workload, and poor real-time performance.

[0003] In summary, the existing online monitoring of partial discharge of high-voltage switchgear has certain limitations and cannot achieve real-time and accurate monitoring of partial discharge of high-voltage switchgear. Therefore, it is of great significance to study an online monitoring method for partial discharge of high-voltage switchgear. Summary of the invention

[0004] The purpose of this application is to provide a method and system for online monitoring of partial discharge of a high-voltage switch cabinet, so as to solve the technical problems in the prior art that the offline monitoring method has a long cycle, a large workload, and poor real-time performance, which leads to the inability to timely detect and handle partial discharge defects, thereby increasing the risk of equipment failure.

[0005] In view of the above problems, the present application provides a method and system for online monitoring of partial discharge of a high-voltage switch cabinet.

[0006] In a first aspect, the present application provides a method for online monitoring of partial discharge of a high-voltage switch cabinet, the method being implemented by an online monitoring system for partial discharge of a high-voltage switch cabinet, wherein the method comprises: performing structural division processing on the high-voltage switch cabinet to obtain a structural part set, wherein the structural part set comprises multiple parts; calling a predetermined key part decision strategy to analyze the multiple parts, and screening to determine the key part set, wherein the predetermined key part decision strategy comprises a first decision strategy and a second decision strategy, wherein the first decision strategy refers to a strategy for determining the key part set in combination with an expert knowledge base, and the second decision strategy refers to a strategy for determining the key part set in combination with a partial discharge database strategy; randomly extract any key part concentrated in the key parts, and record it as the target part; obtain the target partial discharge signal by dynamically monitoring the target partial discharge sensor arranged at the target part; input the target partial discharge feature of the target partial discharge signal into the partial discharge classifier, and obtain the target partial discharge judgment result through the partial discharge classifier, wherein the target partial discharge feature refers to the partial discharge feature obtained by collecting the features of the pre-processed target partial discharge signal; according to the target partial discharge judgment result, if the target part meets the predetermined conditions, issue an early warning instruction; according to the early warning instruction, a partial discharge early warning is issued for the target part of the high-voltage switchgear.

[0007] In a second aspect, the present application further provides an online partial discharge monitoring system for a high-voltage switch cabinet, which is used to execute an online partial discharge monitoring method for a high-voltage switch cabinet as described in the first aspect, wherein the system includes: an obtaining unit, which is used to perform structural division processing on the high-voltage switch cabinet to obtain a set of structural parts, wherein the set of structural parts includes multiple parts; a determining unit, which is used to call a predetermined key part decision strategy to analyze the multiple parts, and screen and determine the key part set, wherein the predetermined key part decision strategy includes a first decision strategy and a second decision strategy, wherein the first decision strategy refers to a strategy for determining the key part set in combination with an expert knowledge base, and the second decision strategy refers to a strategy for determining the key part set in combination with a partial discharge database; an extracting unit, which uses A method for randomly extracting any key part concentrated in the key parts and recording it as a target part; a second obtaining unit is used to obtain a target partial discharge signal by dynamically monitoring through a target partial discharge sensor arranged at the target part; a third obtaining unit is used to input a target partial discharge feature of the target partial discharge signal into a partial discharge classifier, and obtain a target partial discharge judgment result through the partial discharge classifier, wherein the target partial discharge feature refers to a partial discharge feature obtained by collecting features of the pre-processed target partial discharge signal; an issuing unit is used to issue a warning instruction according to the target partial discharge judgment result if the target part meets a predetermined condition; an execution unit is used to issue a partial discharge warning for the target part of the high-voltage switchgear according to the warning instruction.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] The structure part set is obtained by performing structural division processing on the high-voltage switch cabinet, and the structure part set includes multiple parts; the predetermined key part decision strategy is called to analyze the multiple parts, and the key part set is screened and determined, wherein the predetermined key part decision strategy includes a first decision strategy and a second decision strategy, wherein the first decision strategy refers to a strategy for determining the key part set in combination with an expert knowledge base, and the second decision strategy refers to a strategy for determining the key part set in combination with a partial discharge database; any key part in the key part set is randomly extracted and recorded as a target part; a target partial discharge signal is obtained by dynamic monitoring of a target partial discharge sensor arranged at the target part; the target partial discharge feature of the target partial discharge signal is input into a partial discharge classifier, and a target partial discharge judgment result is obtained by the partial discharge classifier, wherein the target partial discharge feature refers to a partial discharge feature obtained by collecting features of the pre-processed target partial discharge signal; according to the target partial discharge judgment result, if the target part meets the predetermined conditions, an early warning instruction is issued; and a partial discharge early warning is performed on the target part of the high-voltage switch cabinet according to the early warning instruction. By adopting an online monitoring method, real-time monitoring of partial discharge of the high-voltage switch cabinet is realized, and the real-time performance of the monitoring system is improved.

[0010] 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 according to the contents of the specification, and 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 specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 A schematic diagram of a flow chart of an online partial discharge monitoring method for a high-voltage switch cabinet according to the present application;

[0013] Figure 2 This is a structural schematic diagram of a high-voltage switchgear partial discharge online monitoring system for the present application.

[0014] Description of reference numerals:

[0015] An obtaining unit 11 , a determining unit 12 , an extracting unit 13 , a second obtaining unit 14 , a third obtaining unit 15 , an issuing unit 16 , and an executing unit 17 . DETAILED DESCRIPTION

[0016] The present application provides a method and system for online monitoring of partial discharge of high-voltage switch cabinets, which solves the problem that online monitoring of partial discharge of high-voltage switch cabinets in the prior art has certain limitations and cannot realize real-time and accurate monitoring of partial discharge of high-voltage switch cabinets. The method realizes real-time monitoring of partial discharge of high-voltage switch cabinets and improves the real-time performance of the monitoring system.

[0017] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0018] For example, please refer to the attached Figure 1 The present application provides a method for online monitoring of partial discharge of a high-voltage switch cabinet, wherein the method is applied to an online monitoring system for partial discharge of a high-voltage switch cabinet, and the method specifically comprises the following steps:

[0019] Step P10: performing structural division processing on the high-voltage switch cabinet to obtain a structural part set, wherein the structural part set includes a plurality of parts;

[0020] Step P20: calling a predetermined key part decision strategy to analyze the multiple parts, and screening to determine a key part set, wherein the predetermined key part decision strategy includes a first decision strategy and a second decision strategy, the first decision strategy refers to a strategy for determining the key part set in combination with an expert knowledge base, and the second decision strategy refers to a strategy for determining the key part set in combination with a partial discharge database;

[0021] Step P30: randomly extracting any key part from the key part set and recording it as the target part;

[0022] Step P40: obtaining a target partial discharge signal through dynamic monitoring by a target partial discharge sensor arranged at the target location;

[0023] Step P50: inputting the target partial discharge feature of the target partial discharge signal into a partial discharge classifier, and obtaining a target partial discharge judgment result through the partial discharge classifier, wherein the target partial discharge feature refers to a partial discharge feature obtained by collecting features of the pre-processed target partial discharge signal;

[0024] Step P60: Based on the target partial discharge judgment result, if the target part meets the predetermined conditions, a warning instruction is issued;

[0025] Step P70: Performing a partial discharge warning on the target part of the high-voltage switchgear according to the warning instruction.

[0026] Specifically, through structural division processing, key part decision-making strategy, and dynamic monitoring and classification, effective monitoring and early warning of local discharge in key parts of high-voltage switchgear are achieved, which not only improves the safety performance of the equipment, but also facilitates the early detection and handling of potential faults, ultimately ensuring the stable operation of the power system.

[0027] First, the high-voltage switchgear is structurally divided to obtain a set of structural parts including multiple parts. Then, the predetermined key part decision strategy is called to analyze these parts, screen and determine the key part set. Next, a key part is randomly extracted from the key part set as the target part. A partial discharge sensor is arranged at the target part to dynamically monitor and obtain the target partial discharge signal. The characteristics of this signal are input into the partial discharge classifier, and the target partial discharge judgment result is obtained through classifier analysis. According to this judgment result, if the target part meets the predetermined conditions, an early warning instruction is issued. Finally, a partial discharge early warning is issued for the target part of the high-voltage switchgear according to the early warning instruction.

[0028] In addition, the decision-making strategy for key parts and the selection of partial discharge classifiers are crucial to the accuracy and efficiency of monitoring. Among them, the selection of key parts needs to be based on a deep understanding of the high-voltage switchgear structure and the identification of fault modes. The training and optimization of partial discharge classifiers require a large amount of effective partial discharge signal data. In addition, the issuance and response mechanism of early warning instructions also need to be closely integrated with the existing maintenance process to ensure that the early warning can be handled in a timely and effective manner.

[0029] Through the above steps, the partial discharge situation of the key parts of the high-voltage switchgear can be monitored and warned in real time, thus greatly improving the safety of the equipment and the reliability of the power system. Obviously, this method has a significant effect on preventing equipment failures and accidents caused by partial discharge.

[0030] The method further comprises the steps of:

[0031] Extract the first decision strategy among the predetermined key part decision strategies; introduce an expert knowledge base according to the first decision strategy, and obtain multiple critical weight coefficients of the multiple parts in sequence based on the expert knowledge base; arrange the multiple parts in descending order based on the multiple critical weight coefficients to obtain a part descending list; add the parts with a predetermined ranking threshold in the part descending list to the key part set.

[0032] Specifically, the expert knowledge base and critical weight coefficients are used to screen the key parts of the high-voltage switchgear. By extracting the first decision strategy, introducing the expert knowledge base, and assigning critical weight coefficients to each part, and then sorting these parts, the key part set is finally determined. This method can ensure that the selection of key parts is more scientific and accurate, thereby improving the effectiveness and reliability of the partial discharge early warning system.

[0033] First, extract the first decision strategy among the predetermined key parts decision strategies. Then, according to this strategy, introduce the expert knowledge base, which contains the experience and knowledge of experts in the field. Next, based on the expert knowledge base, analyze multiple parts of the high-voltage switchgear and obtain their critical weight coefficients in turn. These weight coefficients reflect the importance of each part in partial discharge warning. Then, based on these weight coefficients, arrange multiple parts in descending order to form a descending list of parts. Finally, according to the predetermined ranking threshold, add the top-ranked parts in the descending list of parts to the key part set. In addition, the construction of the expert knowledge base and the determination of the weight coefficient are the key to this process. The expert knowledge base should contain rich historical data and expert experience to ensure the accuracy of the weight coefficient. In addition, the setting of the ranking threshold also needs to be determined based on the actual application scenario and safety requirements to ensure that the key part set can cover all important safety risk points.

[0034] Through the above steps, the expert knowledge base and critical weight coefficients are effectively utilized to achieve scientific screening and sorting of key parts of high-voltage switchgear. Obviously, this method can improve the pertinence and accuracy of partial discharge warning, and help to more effectively prevent and deal with potential faults caused by partial discharge.

[0035] The method further comprises the steps of:

[0036] Extracting a second decision strategy from the predetermined key part decision strategies; retrieving a partial discharge database according to the second decision strategy, wherein the partial discharge database includes multiple groups of partial discharge records of the multiple parts; extracting a first part from the multiple parts, and matching a first partial discharge record group of the first part in the multiple groups of partial discharge records; when the partial discharge frequency in the first partial discharge record group reaches a predetermined frequency threshold, adding the first part to the key part set.

[0037] Specifically, the key parts of the high-voltage switchgear are further screened through the partial discharge database and the second decision strategy. Multiple sets of partial discharge records in the partial discharge database are used to analyze each part, with special attention paid to the partial discharge frequency. By matching the partial discharge records of a specific part and judging whether its partial discharge frequency reaches a predetermined threshold, it is determined whether the part is a key part.

[0038] First, extract the second decision strategy in the predetermined key part decision strategy. Then, according to this strategy, retrieve the partial discharge database, which contains multiple groups of partial discharge records for multiple parts. Next, extract the first part of the multiple parts, and find the first partial discharge record group corresponding to this part in the partial discharge database. Then, analyze the partial discharge frequency in this partial discharge record group. When this frequency reaches the predetermined frequency threshold, it means that the partial discharge activity of this part is frequent and there may be a high risk of failure. Therefore, this part is added to the key part set. In addition, the construction and maintenance of the partial discharge database is the key to this process. The database needs to be updated regularly to ensure the accuracy and timeliness of the partial discharge records. In addition, the setting of the predetermined frequency threshold also needs to be determined based on actual operating data and safety requirements to ensure that key parts are not missed or misreported.

[0039] Through the above steps, the technical solution effectively utilizes the partial discharge database and the second decision-making strategy to achieve accurate screening of key parts of the high-voltage switchgear. Obviously, this method can improve the accuracy and practicality of partial discharge warning, and help to more effectively identify and prevent potential faults caused by partial discharge.

[0040] The method further comprises the steps of:

[0041] The target partial discharge signal is filtered to obtain a target filtered signal; the target filtered signal is amplified to obtain a target amplified signal; the target amplified signal is demodulated to obtain a target pre-processed partial discharge signal; a predetermined partial discharge feature is acquired, and based on the predetermined partial discharge feature, feature collection is performed on the target pre-processed partial discharge signal to obtain the target partial discharge feature.

[0042] Specifically, a series of processing and feature extraction are performed on the target PD signal to facilitate subsequent PD classification and early warning. The pre-processed PD signal is obtained by filtering, amplifying and demodulating the original PD signal, and then based on the predetermined PD feature, features are collected from the pre-processed signal to finally obtain the target PD feature.

[0043] First, the target partial discharge signal is filtered to remove noise and interference in the signal to obtain the target filtered signal. Then, the target filtered signal is amplified to enhance the signal strength and obtain the target amplified signal. Next, the target amplified signal is demodulated. This step helps to extract the envelope information of the signal and obtain the target pre-processed partial discharge signal. In addition, the predetermined partial discharge features are obtained. These features are pre-set according to the characteristics of the partial discharge signal and the classification requirements. Finally, based on these predetermined partial discharge features, the target pre-processed partial discharge signal is feature collected to obtain the target partial discharge features. In addition, the selection of parameters for filtering, amplification and demodulation is crucial to the signal processing effect. These parameters need to be determined based on the actual partial discharge signal characteristics and system requirements to ensure the effectiveness and accuracy of signal processing. At the same time, the predetermined partial discharge features also need to be selected according to the actual application scenarios and classification requirements to ensure the effectiveness of the features and the accuracy of the classification.

[0044] Through the above steps, the technical solution achieves effective processing and feature extraction of the target partial discharge signal. Obviously, this method can improve the accuracy and effectiveness of partial discharge signal processing and provide reliable data support for subsequent partial discharge classification and early warning.

[0045] The method further comprises the following steps: the predetermined partial discharge characteristics at least include discharge times, discharge phases and discharge amplitudes.

[0046] Specifically, three main predetermined partial discharge characteristics are determined: discharge times, discharge phase and discharge amplitude. These three characteristics are important indicators for evaluating partial discharge activities, and can provide key information about the frequency, timing and intensity of partial discharge. Through these characteristics, the severity and potential risks of partial discharge can be analyzed and judged more accurately, thereby providing strong data support for partial discharge early warning. By determining the discharge times, discharge phase and discharge amplitude as predetermined partial discharge characteristics, the accuracy and comprehensiveness of partial discharge analysis and early warning can be improved.

[0047] The method further comprises the steps of:

[0048] A first sample data group is formed based on the partial discharge database; a non-partial discharge database is retrieved, and a second sample data group is formed based on the non-partial discharge database; the first sample data group and the second sample data group are learned based on the support vector machine principle to obtain the partial discharge classifier.

[0049] Specifically, by constructing and training a sample data set, a partial discharge classifier is obtained using the principle of support vector machine. First, a first sample data set is formed based on the partial discharge database, which contains known partial discharge conditions. Then, the non-partial discharge database is retrieved to form a second sample data set, which represents the normal operating state. By comparing and learning these two sample data sets, a partial discharge classifier that can distinguish between partial discharge and normal state can be trained. This method can improve the accuracy and efficiency of partial discharge identification and provide reliable technical support for partial discharge early warning.

[0050] First, the first sample data group is formed based on the partial discharge database. These data should cover a variety of partial discharge situations to ensure that the classifier can identify a variety of partial discharge modes. Then, the non-partial discharge database is retrieved to form the second sample data group. These data should represent the state of the equipment during normal operation and serve as a reference for the normal state. Next, the first sample data group and the second sample data group are learned based on the support vector machine principle. Support vector machine is an effective classification algorithm that can find the best hyperplane to distinguish different data categories. In addition, the construction and learning process of the sample data group is crucial to the performance of the classifier. The sample data needs to be representative, diverse, and sufficient in number to ensure that the classifier can accurately identify partial discharge and normal states. At the same time, the parameter selection and model optimization of the support vector machine also need to be determined based on the actual data and classification requirements to ensure the effectiveness and accuracy of the classifier.

[0051] Through the above steps, a sample data set is constructed based on the partial discharge database and the non-partial discharge database, and a partial discharge classifier is obtained using the support vector machine principle.

[0052] The method further comprises the steps of:

[0053] Extract any partial discharge record group in the partial discharge database; obtain any partial discharge feature in the arbitrary partial discharge record group, and form the first sample data group with the partial discharge label.

[0054] Specifically, a first sample data group is constructed for training a partial discharge classifier. The first sample data group is formed by extracting any partial discharge record group in the partial discharge database and obtaining any partial discharge features in these record groups, and then combining these features with the corresponding partial discharge labels. Such a sample data group contains the specific features of partial discharge and its classification labels, and is the basis for training a partial discharge classifier. In this way, it can be ensured that the classifier can learn and identify the characteristics of partial discharge, thereby improving the accuracy and reliability of partial discharge warning. The construction of the sample data group is crucial to the training and performance of the classifier. The selection of partial discharge record groups needs to be representative and diverse to ensure that the sample data can fully reflect various situations of partial discharge. At the same time, the extraction of partial discharge features and the accuracy of labels also need to be guaranteed to ensure the quality of the sample data and the effectiveness of the classifier.

[0055] The method further comprises the steps of:

[0056] Extract any non-partial discharge record group in the non-partial discharge database; obtain any non-partial discharge feature in the any non-partial discharge record group, and form the second sample data group with the non-partial discharge label.

[0057] Specifically, a second sample data group is constructed for training a partial discharge classifier to distinguish normal states. By extracting any non-partial discharge record group from the non-partial discharge database, and obtaining any non-partial discharge features in these record groups, these features are then combined with corresponding non-partial discharge labels to form a second sample data group. Such a sample data group contains the features and classification labels of the equipment under normal operating conditions, which is crucial for training the partial discharge classifier to identify normal states. In this way, it can be ensured that the classifier can learn and distinguish the features of normal states, thereby improving the accuracy and reliability of partial discharge warning. First, any non-partial discharge record group is extracted from the non-partial discharge database. These record groups should contain information under normal operating conditions of the equipment to facilitate subsequent feature extraction and classification. Then, any non-partial discharge features are obtained from these non-partial discharge record groups. These features may include normal discharge times, discharge phases, discharge amplitudes, etc., which are key indicators for evaluating the normal operating conditions of the equipment. Next, these non-partial discharge features are combined with corresponding non-partial discharge labels to form a second sample data group. The non-partial discharge label is the classification information corresponding to each non-partial discharge feature, which is used to indicate whether the feature belongs to the normal operating state of the equipment.

[0058] The method further comprises the steps of:

[0059] The target part is continuously monitored by an ultrasonic sensor to obtain a target ultrasonic signal time series, wherein the ultrasonic sensor is arranged at the target part; the target ultrasonic signal time series is decomposed to obtain a decomposition result, wherein the decomposition result includes multiple components; characteristic evaluation is performed on the multiple components, and the target component is screened to determine the target component; a target characteristic parameter set of the target component is weightedly calculated to obtain a target component coefficient; and the target component coefficient is used to characterize the severity of the partial discharge of the target part.

[0060] Specifically, ultrasonic sensors are used to continuously monitor the target parts of the high-voltage switchgear, and the obtained ultrasonic signal timing is used to evaluate the severity of partial discharge. Ultrasonic sensors are placed at the target parts, ultrasonic signals are collected, and then the signal timing is decomposed to obtain multiple components. Then, these components are characterized, the target components are screened out, and their characteristic parameter sets are weighted to obtain the target component coefficient. This coefficient is used to characterize the severity of partial discharge at the target part, thereby providing an important evaluation basis for partial discharge early warning.

[0061] First, the target part is continuously monitored by ultrasonic sensors to obtain the target ultrasonic signal time series. The location and number of ultrasonic sensors need to be determined according to the characteristics of the target part and the monitoring requirements to ensure that the ultrasonic signals generated by partial discharge can be effectively collected. Then, the target ultrasonic signal time series is decomposed, such as using methods such as Hilbert-Huang Transform (HHT) to obtain a decomposition result containing multiple components. These components represent the characteristics of the signal at different frequencies and time periods.

[0062] Next, multiple components are evaluated for characteristics, and the target component is screened and determined. The basis for characteristic evaluation can be indicators such as the energy, frequency, and stability of the component. The selection of the target component needs to be determined based on the actual application scenario and partial discharge characteristics. Then, the target characteristic parameter set of the target component is weighted and the target component coefficient is obtained. The weight of the weighted calculation needs to be determined according to the importance of different parameters and the actual application requirements.

[0063] Finally, the target component coefficient is used to characterize the severity of partial discharge at the target location. This coefficient reflects the intensity of partial discharge and the possible degree of damage, and can serve as an important basis for partial discharge early warning.

[0064] Through the above steps, the partial discharge severity assessment of the target part of the high-voltage switchgear is achieved.

[0065] In summary, the method for online monitoring of partial discharge of a high-voltage switch cabinet provided by the present application has the following technical effects:

[0066] 1. A structure part set is obtained by structurally dividing the high-voltage switchgear, and the structure part set includes multiple parts; a predetermined key part decision strategy is called to analyze the multiple parts, and a key part set is screened and determined, wherein the predetermined key part decision strategy includes a first decision strategy and a second decision strategy, wherein the first decision strategy refers to a strategy for determining the key part set in combination with an expert knowledge base, and the second decision strategy refers to a strategy for determining the key part set in combination with a partial discharge database; any key part in the key part set is randomly extracted and recorded as a target part; a target partial discharge signal is obtained by dynamically monitoring the target partial discharge sensor arranged at the target part; the target partial discharge feature of the target partial discharge signal is input into a partial discharge classifier, and a target partial discharge judgment result is obtained by the partial discharge classifier, wherein the target partial discharge feature refers to a partial discharge feature obtained by collecting features of the pre-processed target partial discharge signal; according to the target partial discharge judgment result, if the target part meets the predetermined conditions, an early warning instruction is issued; and a partial discharge early warning is performed on the target part of the high-voltage switchgear according to the early warning instruction. By adopting an online monitoring method, real-time monitoring of partial discharge of the high-voltage switchgear is realized, and the real-time performance of the monitoring system is improved.

[0067] 2. Use advanced signal recognition technology to extract and classify the features of partial discharge signals, achieve accurate recognition of partial discharge signals, and improve the reliability of the monitoring system.

[0068] 3. Through real-time detection and analysis of ultrasonic signals, the severity of partial discharge is intelligently predicted, which is of great significance for quantifying partial discharge.

[0069] Embodiment 2: Based on the same inventive concept as the method for online monitoring partial discharge of a high-voltage switch cabinet in the above embodiment, the present application also provides an online monitoring system for partial discharge of a high-voltage switch cabinet, as shown in the attached Figure 2 , the system comprising:

[0070] The obtaining unit 11 is used to perform structural division processing on the high-voltage switch cabinet to obtain a set of structural parts, wherein the set of structural parts includes multiple parts; the determining unit 12 is used to call a predetermined key part decision strategy to analyze the multiple parts, and screen and determine the key part set, wherein the predetermined key part decision strategy includes a first decision strategy and a second decision strategy, wherein the first decision strategy refers to a strategy for determining the key part set in combination with an expert knowledge base, and the second decision strategy refers to a strategy for determining the key part set in combination with a partial discharge database; the extracting unit 13 is used to randomly extract any key part in the key part set and record it as a target part; the second obtaining unit Element 14 is used to obtain a target partial discharge signal by dynamically monitoring the target partial discharge sensor arranged at the target part; a third obtaining unit 15 is used to input the target partial discharge feature of the target partial discharge signal into a partial discharge classifier, and obtain a target partial discharge judgment result through the partial discharge classifier, wherein the target partial discharge feature refers to the partial discharge feature obtained by collecting features of the pre-processed target partial discharge signal; an issuing unit 16 is used to issue an early warning instruction according to the target partial discharge judgment result if the target part meets a predetermined condition; an execution unit 17 is used to issue a partial discharge early warning for the target part of the high-voltage switchgear according to the early warning instruction.

[0071] Furthermore, the determination unit 12 in the system is also used to extract a first decision strategy from the predetermined key part decision strategies; introduce an expert knowledge base according to the first decision strategy, and obtain multiple critical weight coefficients of the multiple parts in sequence based on the expert knowledge base; arrange the multiple parts in descending order based on the multiple critical weight coefficients to obtain a part descending list; and add parts with a predetermined ranking threshold in the part descending list to the key part set.

[0072] Furthermore, the determination unit 12 in the system is also used to extract a second decision strategy from the predetermined key part decision strategy; retrieve a partial discharge database according to the second decision strategy, wherein the partial discharge database includes multiple groups of partial discharge records for the multiple parts; extract a first part from the multiple parts, and match a first partial discharge record group of the first part in the multiple groups of partial discharge records; when the partial discharge frequency in the first partial discharge record group reaches a predetermined frequency threshold, add the first part to the key part set.

[0073] Furthermore, the third obtaining unit 15 in the system is also used to filter the target partial discharge signal to obtain a target filtered signal; amplify the target filtered signal to obtain a target amplified signal; demodulate the target amplified signal to obtain a target pre-processed partial discharge signal; obtain a predetermined partial discharge feature, and collect features of the target pre-processed partial discharge signal based on the predetermined partial discharge feature to obtain the target partial discharge feature.

[0074] Furthermore, the third obtaining unit 15 in the system is also used for the predetermined partial discharge characteristics to at least include discharge times, discharge phases and discharge amplitudes.

[0075] Furthermore, the determination unit 12 in the system is also used to form a first sample data group based on the partial discharge database; retrieve a non-partial discharge database, and form a second sample data group based on the non-partial discharge database; and learn the first sample data group and the second sample data group based on the support vector machine principle to obtain the partial discharge classifier.

[0076] Furthermore, the determination unit 12 in the system is further configured to extract any partial discharge record group in the partial discharge database; obtain any partial discharge feature in the arbitrary partial discharge record group, and form the first sample data group with the partial discharge label.

[0077] Furthermore, the determination unit 12 in the system is further configured to extract any non-partial discharge record group in the non-partial discharge database; obtain any non-partial discharge feature in the any non-partial discharge record group, and form the second sample data group with the non-partial discharge label.

[0078] Furthermore, the system also includes an evaluation unit, which is used to continuously monitor the target part through an ultrasonic sensor to obtain a target ultrasonic signal time series, wherein the ultrasonic sensor is arranged at the target part; decompose the target ultrasonic signal time series to obtain a decomposition result, and the decomposition result includes multiple components; perform feature evaluation on the multiple components, and screen and determine the target component; perform weighted calculation on the target feature parameter set of the target component to obtain a target component coefficient; and use the target component coefficient to characterize the severity of partial discharge at the target part.

[0079] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1The method and specific examples of online monitoring of partial discharge of a high-voltage switch cabinet in the first embodiment are also applicable to an online monitoring system of partial discharge of a high-voltage switch cabinet in the present embodiment. Through the above detailed description of the method of online monitoring of partial discharge of a high-voltage switch cabinet, those skilled in the art can clearly know the online monitoring system of partial discharge of a high-voltage switch cabinet in the present embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0080] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0081] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A method for online monitoring of partial discharge of a high-voltage switch cabinet, characterized in that: The method comprises: Performing structural division processing on the high-voltage switch cabinet to obtain a structural part set, wherein the structural part set includes a plurality of parts; Calling a predetermined key part decision strategy to analyze the multiple parts, and screening to determine a key part set, wherein the predetermined key part decision strategy includes a first decision strategy and a second decision strategy, the first decision strategy refers to a strategy for determining the key part set in combination with an expert knowledge base, and the second decision strategy refers to a strategy for determining the key part set in combination with a partial discharge database; Randomly extract any key part from the key part set and record it as the target part; A target partial discharge signal is obtained by dynamic monitoring of a target partial discharge sensor arranged at the target location; Inputting the target partial discharge feature of the target partial discharge signal into a partial discharge classifier, and obtaining a target partial discharge judgment result through the partial discharge classifier, wherein the target partial discharge feature refers to a partial discharge feature obtained by collecting features of the preprocessed target partial discharge signal; According to the target partial discharge judgment result, if the target part meets the predetermined conditions, a warning instruction is issued; A partial discharge warning is performed on the target part of the high-voltage switch cabinet according to the warning instruction.

2. The method according to claim 1, characterized in that: The predetermined key part decision strategy is called to analyze the multiple parts, and the key part set is screened and determined, including: Extracting a first decision strategy from the predetermined key part decision strategies; Introducing an expert knowledge base according to the first decision-making strategy, and sequentially obtaining a plurality of critical weight coefficients of the plurality of parts based on the expert knowledge base; Arranging the multiple parts in descending order based on the multiple critical weight coefficients to obtain a descending list of parts; Add the parts with a predetermined ranking threshold in the descending list of parts to the key part set.

3. The method according to claim 2, characterized in that: Adding the parts with a predetermined ranking threshold in the descending list of parts to the key part set further includes: Extracting a second decision strategy from the predetermined key part decision strategies; Retrieving a partial discharge database according to the second decision strategy, wherein the partial discharge database includes multiple groups of partial discharge records at the multiple locations; Extracting a first portion from the multiple portions, and matching a first partial discharge record group of the first portion in the multiple groups of partial discharge records; When the partial discharge frequency in the first partial discharge record group reaches a predetermined frequency threshold, the first part is added to the key part set.

4. The method according to claim 1, characterized in that: Inputting the target partial discharge feature of the target partial discharge signal into a partial discharge classifier, and obtaining a target partial discharge judgment result through the partial discharge classifier, including: Performing filtering processing on the target partial discharge signal to obtain a target filtered signal; amplifying the target filtered signal to obtain a target amplified signal; Demodulating the target amplified signal to obtain a target pre-processed partial discharge signal; A predetermined partial discharge feature is acquired, and based on the predetermined partial discharge feature, feature collection is performed on the target pre-processed partial discharge signal to obtain the target partial discharge feature.

5. The method according to claim 4, characterized in that: The predetermined partial discharge characteristics at least include discharge times, discharge phases and discharge amplitudes.

6. The method according to claim 3, characterized in that: Inputting the target partial discharge feature of the target partial discharge signal into a partial discharge classifier, and obtaining a target partial discharge judgment result through the partial discharge classifier, including: Establishing a first sample data group based on the partial discharge database; Retrieving a non-partial discharge database, and forming a second sample data group according to the non-partial discharge database; The first sample data group and the second sample data group are learned based on the support vector machine principle to obtain the partial discharge classifier.

7. The method according to claim 6, characterized in that: A first sample data group is formed based on the partial discharge database, comprising: Extracting any partial discharge record group in the partial discharge database; Obtain any partial discharge feature in the arbitrary partial discharge record group, and form the first sample data group with the partial discharge label.

8. The method according to claim 6, characterized in that: Retrieving a non-partial discharge database and forming a second sample data group according to the non-partial discharge database, including: Extracting any non-partial discharge record group in the non-partial discharge database; Obtain any non-partial discharge feature in the any non-partial discharge record group, and form the second sample data group with the non-partial discharge label.

9. The method according to claim 1, characterized in that: Performing a partial discharge warning on the target part of the high-voltage switch cabinet according to the warning instruction also includes: Continuously monitoring the target part by an ultrasonic sensor to obtain a target ultrasonic signal timing sequence, wherein the ultrasonic sensor is arranged at the target part; Decomposing the target ultrasonic signal time series to obtain a decomposition result, wherein the decomposition result includes multiple components; Performing feature evaluation on the multiple components and screening to determine the target component; Performing weighted calculation on a target characteristic parameter set of the target component to obtain a target component coefficient; The target component coefficient is used to characterize the severity of the partial discharge at the target site.

10. A high-voltage switchgear partial discharge online monitoring system, characterized in that: The system is used to perform the steps of the method according to any one of claims 1 to 9, and the system comprises: An obtaining unit is used to perform structural division processing on the high-voltage switch cabinet to obtain a structural part set, wherein the structural part set includes a plurality of parts; A determination unit, used for calling a predetermined key part decision strategy to analyze the multiple parts, and screening and determining a key part set, wherein the predetermined key part decision strategy includes a first decision strategy and a second decision strategy, the first decision strategy refers to a strategy for determining the key part set in combination with an expert knowledge base, and the second decision strategy refers to a strategy for determining the key part set in combination with a partial discharge database; An extraction unit, used for randomly extracting any key part from the key part set and recording it as a target part; A second obtaining unit is used to obtain a target partial discharge signal through dynamic monitoring by a target partial discharge sensor arranged at the target part; A third obtaining unit is used to input the target partial discharge feature of the target partial discharge signal into a partial discharge classifier, and obtain a target partial discharge judgment result through the partial discharge classifier, wherein the target partial discharge feature refers to a partial discharge feature obtained by collecting features of the preprocessed target partial discharge signal; An issuing unit is used to issue a warning instruction according to the target partial discharge judgment result if the target part meets the predetermined conditions; An execution unit is used to perform a partial discharge warning on the target part of the high-voltage switch cabinet according to the warning instruction.

Images