GIS partial discharge fault detection device and method based on D-S evidence multi-source information fusion

Abstract

The invention discloses a GIS partial discharge fault detection device and method based on D-S evidence multi-source information fusion, and relates to the technical field of power system control, and the device comprises a signal collection module which is used for collecting an electromagnetic wave signal, a mechanical vibration signal, an ultrasonic signal and a partial discharge optical radiation signal generated by the partial discharge in GIS equipment; the GIS partial discharge fault detection module is used for processing, storing and sending the signals acquired by the signal acquisition module; the coordinator is used for realizing data transmission between the GIS partial discharge fault detection module and the background terminal; and the background terminal is used for receiving the processing information of the GIS partial discharge fault detection module and judging the fault type of the GIS equipment. According to the GIS partial discharge fault diagnosis system, various signals generated by partial discharge in the GIS equipment are detected through the information acquisition module, and the relationship between the characteristic data and the GIS partial discharge fault is established, so that diagnosis of the GIS partial discharge fault is realized, and the fault judgment efficiency of the background terminal is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical equipment detection, and in particular to a GIS partial discharge fault detection device and method based on DS evidence multi-source information fusion. Background Art

[0002] With the extensive application of gas insulated switchgear (GIS) in the fields of ultra-high voltage and ultra-high voltage, the complexity of power equipment and the correlation between equipment have increased, which has put forward higher requirements for the safety of GIS equipment. However, in recent years, "old" GIS equipment has frequently failed, among which the insulation defects of equipment are mainly manifested in the form of partial discharge (PD), which always endangers the safe operation of the power transmission and transformation network. Using artificial intelligence methods to conduct deep learning and analysis of GIS equipment operation data is the development trend in the field of intelligent operation and maintenance of power grids at this stage. Accurately grasping the operation status of equipment is of great significance to ensure the stable operation of the power grid.

[0003] Currently known GIS partial discharge fault detection technologies are as follows:

[0004] (1) The use of fluorescent fiber optic sensors to detect the optical signals generated by PD inside the equipment to diagnose insulation faults has shown obvious advantages. However, due to the blind spot of optical sensors, when there are obstructions between the sensor and the light source, the detection effect of the optical sensor will be seriously affected, and even the sensor will not work properly.

[0005] (2) The ultra high frequency (UHF) detection method is a relatively mature partial discharge detection technology. However, there is electromagnetic interference in the substation environment, and the detected UHF signal contains noise. Its disadvantage is that this method is difficult to implement when there is severe electromagnetic interference in the environment.

[0006] (3) When there are insulation defects inside GIS, the local discharge caused by the breakdown of the insulating medium is often accompanied by the emission of ultrasonic waves. Therefore, ultrasonic local discharge detection has good anti-interference ability and is insensitive to electrical interference, but is easily affected by mechanical or electromagnetic vibrations. It has high detection sensitivity for free particle defects, but low sensitivity for defects on the surface and inside of solid insulation. It can detect non-discharge defects such as loose spring pads and flying dust. The propagation attenuation is large and the detection range is small, which is suitable for defect location.

[0007] In summary, the current detection of GIS operating status is mostly limited to a single physical quantity detection method. The existing detection system can only measure and analyze a single electrical or mechanical state of the equipment, making it difficult to achieve comprehensive detection and evaluation of the operating status of GIS equipment. Summary of the invention

[0008] The technical problem to be solved by the present invention is to solve the deficiencies in the prior art and to design a GIS partial discharge fault detection device and method based on DS evidence multi-source information fusion.

[0009] The technical solution adopted by the present invention to solve its technical problem is:

[0010] In a first aspect, the present invention provides a GIS partial discharge fault detection device based on DS evidence multi-source information fusion, comprising:

[0011] Signal acquisition module, used to collect electromagnetic wave signals, mechanical vibration signals, ultrasonic signals and partial discharge light radiation signals generated by partial discharge inside GIS equipment;

[0012] GIS partial discharge fault detection module, used to process, store and send the signals collected by the signal acquisition module;

[0013] Coordinator, used to realize data transmission between GIS partial discharge fault detection module and background terminal;

[0014] The backend terminal receives the processing information from the GIS partial discharge fault detection module and determines the fault type of the GIS equipment.

[0015] As a further technical solution of the present invention, the signal acquisition module comprises:

[0016] UHF sensor is used to detect electromagnetic wave signals generated by partial discharge inside GIS equipment

[0017] Vibration-ultrasonic fusion sensor, used to detect mechanical vibration signals of GIS equipment and ultrasonic signals generated by internal partial discharge;

[0018] The optical sensor is used to capture the partial discharge light radiation signal and transmit it to the photoelectric conversion unit to realize photoelectric conversion.

[0019] As a further technical solution of the present invention, the GIS partial discharge fault detection module comprises:

[0020] The main controller is used to receive and process input information and run the artificial neural network fault diagnosis model trained by the background terminal.

[0021] A data storage device for storing the trained artificial neural network fault diagnosis model and data information;

[0022] A wireless communication network module, used for transmitting each data to the coordinator through a router;

[0023] Human-machine interface, used for parameter setting and data calling.

[0024] As a further technical solution of the present invention, the vibration-ultrasound fusion sensor is a compression piezoelectric acceleration sensor.

[0025] As a further technical solution of the present invention, the optical sensor uses the Jingbang Technology TP3050_SIPM silicon photomultiplier tube as a photoelectric conversion unit of the fluorescent fiber optic detection method.

[0026] As a further technical solution of the present invention, the coordinator is used to establish a ZigBee network, and the coordinator sends data to a background terminal via a GPRS network.

[0027] In a second aspect, the present invention also provides a GIS partial discharge fault intelligent detection method based on DS evidence multi-source information fusion, comprising:

[0028] The signal acquisition module collects electromagnetic wave signals, mechanical vibration signals, ultrasonic signals and partial discharge light radiation signals generated by partial discharge inside GIS equipment;

[0029] The GIS partial discharge fault detection module processes, stores and sends the signals collected by the signal acquisition module;

[0030] The backend terminal receives the processing information from the GIS partial discharge fault detection module and determines the fault type of the GIS equipment.

[0031] As a further technical solution of the present invention, the GIS partial discharge fault detection module processes, stores and sends the signal collected by the signal acquisition module; specifically, it includes:

[0032] The mechanical vibration signal is processed by wavelet packet transform to obtain the vibration time-frequency analysis feature quantity;

[0033] The UHF signal, ultrasonic signal, and optical signal spectrum are extracted through CNN to obtain ultra-high frequency feature quantities, acoustic feature quantities, and optical feature quantities;

[0034] The vibration time-frequency analysis feature quantity, ultra-high frequency feature quantity, acoustic feature quantity and optical feature quantity are normalized and converted into vector machine input data.

[0035] As a further technical solution of the present invention, the background terminal receives the processing information of the GIS partial discharge fault detection module and determines the fault type of the GIS equipment, specifically including:

[0036] Processing the vector machine input data through the support vector machine to obtain the diagnosis result;

[0037] DS decision fusion is used to determine the type of GIS partial discharge fault based on the diagnosis results.

[0038] As a further technical solution of the present invention, the GIS partial discharge fault types include: normal state, contamination, air gap, suspended electrode, protrusion, mechanical mild vibration, mechanical moderate vibration and mechanical severe vibration.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1 The present invention detects various signals generated by partial discharge inside GIS equipment through the information acquisition module, and can effectively capture the physical and chemical change information caused by the generation and development of partial discharge through acoustic, electrical, optical and other monitoring means, and refers to the different discharge modes and characteristic attributes shown by partial discharge of different defect types to judge the type of equipment failure; establishes the relationship between characteristic data and GIS partial discharge failure, so as to realize the diagnosis of GIS partial discharge failure.

[0041] 2. The present invention integrates data processing into the GIS partial discharge fault detection module, which greatly reduces the amount of data processed by the background terminal and improves the fault judgment efficiency of the background terminal.

[0042] 3. The wireless communication network module is based on ZigBee wireless communication network technology, which is suitable for short-distance data transmission and has the characteristics of low power consumption, low cost, short delay, reliability and security. The data is transmitted to the coordinator through the router, and the human-machine interface is mainly used for parameter setting and data call.

[0043] 4. The coordinator is used to build a ZigBee network, and the coordinator sends data to the background terminal through the GPRS network.

[0044] 5. The present invention adopts DS evidence theory to integrate acoustic, electrical, optical and other sensors to effectively capture the physical and chemical change information generated by partial discharge, and makes intelligent diagnosis based on the different discharge modes and characteristic attributes exhibited by partial discharge of different defect types. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is the control structure diagram of the GIS partial discharge fault detection device based on DS evidence multi-source information fusion proposed by the present invention;

[0046] Figure 2 This is a connection diagram of the GIS partial discharge fault detection device based on DS evidence multi-source information fusion proposed by the present invention;

[0047] Figure 3 This is a structural diagram of the vibration-ultrasound fusion sensor proposed by the present invention;

[0048] Figure 4 The JAMP evaluation circuit board schematic diagram of the optical sensor proposed in the present invention;

[0049] Figure 5 This is a flow chart of the judgment method of the GIS partial discharge fault detection device based on DS evidence multi-source information fusion proposed by the present invention;

[0050] Figure 6 This is a processing flow chart of the GIS partial discharge fault detection module proposed by the present invention;

[0051] Figure 7 A flow chart of a specific embodiment of the present invention;

[0052] Figure 8 This is a flowchart of the background terminal processing proposed by the present invention. DETAILED DESCRIPTION

[0053] The specific implementation of the present invention is described below in conjunction with the accompanying drawings and embodiments:

[0054] It should be noted that the structures, colors, proportions, sizes, etc. illustrated in the drawings of this specification are only used to match the contents disclosed in the specification so that people familiar with this technology can understand and read them, and are not used to limit the conditions under which the present invention can be implemented. Any structural modification, change in proportional relationship or adjustment of size should still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0055] Unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present invention should be understood by people with ordinary skills in the field to which the present disclosure belongs. "First", "second" and similar words used in the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0056] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0057] The present invention can effectively capture the physical and chemical change information caused by the generation and development of partial discharge through acoustic, electrical, optical and other monitoring means, and refer to the different discharge modes and characteristic attributes exhibited by partial discharges of different defect types. The present invention uses an artificial intelligence neural network model to establish the relationship between characteristic data and GIS partial discharge faults, thereby realizing the diagnosis of GIS partial discharge faults and ensuring the reliability of equipment and the safe and stable operation of the power grid.

[0058] Embodiment 1

[0059] like Figure 1 and Figure 2 As shown, the present invention provides a GIS partial discharge fault detection device based on DS evidence multi-source information fusion, comprising:

[0060] The signal acquisition module 101 is used to collect electromagnetic wave signals, mechanical vibration signals, ultrasonic signals and partial discharge light radiation signals generated by partial discharge inside the GIS equipment;

[0061] GIS partial discharge fault detection module 102, used for processing, storing and sending the signals collected by the signal collection module;

[0062] Coordinator 103, used to realize data transmission between GIS partial discharge fault detection module and background terminal;

[0063] The backend terminal 104 receives the processing information of the GIS partial discharge fault detection module and determines the fault type of the GIS equipment.

[0064] In the embodiment of the present invention, the information acquisition module detects various signals generated by partial discharge inside the GIS equipment, and the physical and chemical change information caused by the generation and development of partial discharge can be effectively captured by means of acoustic, electrical, optical and other monitoring means, and the equipment fault type can be determined by referring to the different discharge modes and characteristic attributes exhibited by partial discharge of different defect types. The signals collected by the signal acquisition module are processed by the GIS partial discharge fault detection module, and the collected signals are processed into data information that can be input to the support vector machine, and then sent to the background terminal for fault diagnosis through the wireless communication module and the coordinator. The present invention integrates data processing in the GIS partial discharge fault detection module, which greatly reduces the amount of data processed by the background terminal and improves the fault judgment efficiency of the background terminal.

[0065] In the embodiment of the present invention, the signal acquisition module 101 includes:

[0066] UHF sensor 111 is used to detect electromagnetic wave signals generated by partial discharge inside GIS equipment; UHF sensor is mainly used to detect electromagnetic wave signals generated by partial discharge inside GIS equipment. The antenna of UHF sensor is the carrier of the entire sensor, and the performance of the antenna will directly affect the signal acquisition and subsequent processing. Since the spectrum of partial discharge signals is mainly distributed in 300-1500MHz, a broadband UHF antenna is required as a sensor to monitor partial discharge.

[0067] The vibration-ultrasonic fusion sensor 112 is used to detect the mechanical vibration signal of the GIS equipment and the ultrasonic signal generated by the internal partial discharge; the vibration-ultrasonic fusion sensor is mainly used to detect the mechanical vibration signal of the GIS equipment and the ultrasonic signal generated by the internal partial discharge. The compression piezoelectric acceleration sensor is currently widely used in the mechanical vibration detection of bearings, spacecraft and other equipment. Piezoelectric ultrasonic sensors are currently mostly used in fault detection and positioning of power equipment. The principle is based on the acoustic-electric properties of piezoelectric ceramics. When it is close to the surface of the object to be detected, the ultrasonic wave generated by the sound source causes the internal lattice deformation of the piezoelectric material, resulting in a change in the polarization state and outputting an electrical signal, thereby realizing the detection of ultrasonic signals.

[0068] Based on the similarity of the internal structures of the above-mentioned compression piezoelectric acceleration sensor and the piezoelectric ultrasonic sensor, the present invention proposes a new vibration-ultrasonic fusion sensor based on their respective working principles. Figure 3 As shown: the shell 1121 is made of stainless steel material, a mass backing block shell 1122 is arranged inside the shell 1121, an embedded backing 1123 and tungsten + epoxy resin material are arranged inside the mass backing block shell 1122, a piezoelectric ceramic 1124 is arranged below the mass backing block shell 1122, a matching base 1125 is arranged at the bottom of the piezoelectric ceramic 1124, wherein the matching base is made of aluminum alloy material, an insulating layer 1126 is arranged at the bottom of the shell 1121, wherein epoxy resin paint is coated on the bonding layer.

[0069] The optical sensor 113 is used to capture the partial discharge radiation signal and transmit it to the photoelectric conversion unit to realize photoelectric conversion. The optical sensor uses a fluorescent fiber optic sensor to capture the partial discharge radiation signal and transmit it to the photoelectric conversion unit to realize photoelectric conversion. This device uses the Jingbang Technology TP3050_SIPM silicon photomultiplier tube as the photoelectric conversion unit of the fluorescent fiber optic detection method.

[0070] The main internal circuit components are JAMP evaluation circuit boards equipped with TP3050 chips. The circuit principle is as follows Figure 4As shown. In the JAMP evaluation circuit board, the bias terminal of SIPM is connected to a positive voltage, and after passing through the transimpedance amplifier, the Aout output terminal outputs an electrical pulse signal. The amplifier is powered by a dual power supply with a supply voltage of ±5V. The output signal amplification factor is controlled by adjusting the resistor RF and the matching capacitor CF. Increasing the resistor RF can increase the amplification factor, and the output signal amplitude increases. At the same time, the value of the appropriate capacitor CF needs to be adjusted for matching. This article uses the TP3050 chip, the resistor RF is 910Ω, and the capacitor CF is 0.5pF.

[0071] In the embodiment of the present invention, the GIS partial discharge fault detection module 102 includes:

[0072] The main controller 121 is used to receive and process input information and run the artificial neural network fault diagnosis model trained by the background terminal;

[0073] Data storage 122, used to store the trained artificial neural network fault diagnosis model and data information;

[0074] The wireless communication network module 123 is used to transmit the data to the coordinator through the router;

[0075] Human-machine interface 124 is used for parameter setting and data calling.

[0076] The main controller is used to receive and process input information and run the artificial neural network fault diagnosis model trained by the background terminal; the data storage is used to store the trained artificial neural network fault diagnosis model and data information. The device adopts DS evidence theory and integrates the current popular artificial intelligence algorithm for GIS partial discharge fault diagnosis. The wireless communication network module is based on ZigBee wireless communication network technology, which is suitable for short-distance data transmission and has the characteristics of low power consumption, low cost, short delay, reliability and security. The data is transmitted to the coordinator through the router, and the human-machine interface is mainly used for parameter setting and data call.

[0077] In the embodiment of the present invention, the coordinator is used to establish a ZigBee network, and the coordinator sends data to a background terminal via a GPRS network.

[0078] Embodiment 2

[0079] See also Figure 5 This embodiment proposes a GIS partial discharge fault intelligent detection method based on DS evidence multi-source information fusion, including:

[0080] Step 201, collecting electromagnetic wave signals, mechanical vibration signals, ultrasonic signals and partial discharge light radiation signals generated by partial discharge inside the GIS equipment through a signal acquisition module;

[0081] Step 202, the GIS partial discharge fault detection module processes, stores and sends the signal collected by the signal collection module;

[0082] Step 203: the backend terminal receives the processing information of the GIS partial discharge fault detection module and determines the fault type of the GIS equipment.

[0083] Among them, see Figure 6 and Figure 7 In step 202, the GIS partial discharge fault detection module processes, stores and sends the signal collected by the signal collection module 202; specifically including:

[0084] Step 221, using wavelet packet transform to process the mechanical vibration signal to obtain vibration time-frequency analysis feature quantity;

[0085] Step 222, extracting features of UHF (ultra-high frequency) signals, ultrasonic signals, and optical signal spectra through a convolutional neural network (CNN) to obtain ultra-high frequency feature quantities, acoustic feature quantities, and optical feature quantities;

[0086] Step 223, normalize the vibration time-frequency analysis feature quantity, the ultra-high frequency feature quantity, the acoustic feature quantity, and the optical feature quantity respectively and convert them into vector machine input data.

[0087] See also Figure 8 In step 203, the backend terminal receives the processing information of the GIS partial discharge fault detection module and determines the fault type of the GIS equipment, specifically including:

[0088] Step 231, processing the vector machine input data by the support vector machine to obtain the diagnosis result;

[0089] Step 232, using DS decision fusion to determine the type of GIS partial discharge fault based on the diagnosis results.

[0090] The various variations and specific examples of the device in the aforementioned embodiments are also applicable to the GIS partial discharge fault intelligent detection method based on DS evidence multi-source information fusion in the present embodiment. Through the aforementioned detailed description of the GIS partial discharge fault intelligent detection device based on DS evidence multi-source information fusion, those skilled in the art can clearly understand the GIS partial discharge fault intelligent detection method based on DS evidence multi-source information fusion in the present embodiment. Therefore, for the sake of brevity of the specification, it will not be described in detail here.

[0091] Among them, GIS partial discharge fault types include: normal state, contamination, air gap, suspended electrode, protrusion, mechanical mild vibration, mechanical moderate vibration and mechanical severe vibration.

[0092] The specific implementation steps of the present invention are as follows:

[0093] Install various sensors to collect data in real time;

[0094] Set the relevant parameters of the GIS partial discharge fault intelligent detection device through the human-machine interface / back-end terminal;

[0095] This device collects a total of 4 types of signals, among which the vibration signal is processed by wavelet packet transform, and the UHF transmission signal, ultrasonic signal and optical signal spectrum are extracted by convolutional neural network (CNN).

[0096] The wavelet packet decomposition algorithm is:

[0097]

[0098] Wavelet packet reconstruction algorithm:

[0099]

[0100] In formula (1) and (2): d i,j,n is the i-th wavelet packet coefficient of the n-th node in the i-th layer; h k , g k is the expansion coefficient.

[0101] The purpose of this invention is to prove the effectiveness of the denoising algorithm by wavelet packet transform. The determination of the optimal wavelet basis and the optimal threshold of the algorithm is not discussed in depth. For mechanical vibration and ultrasonic partial discharge signals, the db6 wavelet packet basis is selected as the basis function, and the fusion signal is decomposed into 4 layers of wavelet packets. The threshold function selects a soft threshold function with good overall continuity. The function expression is:

[0102]

[0103] In formula (3), λ is the threshold; sign is the sign function; d i,j is the wavelet packet coefficient. Since the fused signal (vibration signal and ultrasonic signal) has a certain continuity in time, the wavelet coefficient generated by the effective signal has a larger modulus value in the wavelet domain analysis; while Gaussian white noise has no continuity in time, so after the wavelet transform, the noise still shows strong randomness in the wavelet domain, and its corresponding wavelet coefficient is small, so the threshold can be used to achieve denoising;

[0104] After feature normalization, the data is converted into a data set that can participate in model training and input into the support vector machine (SVM);

[0105] Finally, the DS evidence theory is used to integrate the diagnostic results of the four signals and make a comprehensive judgment.

[0106] The present invention adopts the DS evidence theory to integrate the algorithm structure of the current artificial intelligence algorithm for GIS partial discharge fault diagnosis, and adopts the DS evidence theory to integrate the physical and chemical change information generated by the partial discharge of acoustic, electrical, optical and other sensors, and makes intelligent diagnosis according to the different discharge modes and characteristic attributes exhibited by the partial discharge of different defect types. Based on the technology of the present invention, it is possible to extend the development of different types of SF6 electrical equipment fault diagnosis and fault location methods.

[0107] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0108] If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0109] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0110] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A GIS partial discharge fault detection device based on DS evidence multi-source information fusion, characterized by: include: Signal acquisition module, used to collect electromagnetic wave signals, mechanical vibration signals, ultrasonic signals and partial discharge light radiation signals generated by partial discharge inside GIS equipment; GIS partial discharge fault detection module, used to process, store and send the signals collected by the signal acquisition module; Coordinator, used to realize data transmission between GIS partial discharge fault detection module and background terminal; The backend terminal receives the processing information from the GIS partial discharge fault detection module and determines the fault type of the GIS equipment.

2. The GIS partial discharge fault detection device based on DS evidence multi-source information fusion according to claim 1 is characterized in that: The signal acquisition module comprises: UHF sensor is used to detect electromagnetic wave signals generated by partial discharge inside GIS equipment Vibration-ultrasonic fusion sensor, used to detect mechanical vibration signals of GIS equipment and ultrasonic signals generated by internal partial discharge; The optical sensor is used to capture the partial discharge light radiation signal and transmit it to the photoelectric conversion unit to realize photoelectric conversion.

3. The GIS partial discharge fault detection device based on DS evidence multi-source information fusion according to claim 1 is characterized in that: The GIS partial discharge fault detection module comprises: The main controller is used to receive and process input information and run the artificial neural network fault diagnosis model trained by the background terminal. A data storage device for storing the trained artificial neural network fault diagnosis model and data information; A wireless communication network module, used for transmitting each data to the coordinator through a router; Human-machine interface, used for parameter setting and data calling.

4. The GIS partial discharge fault detection device based on DS evidence multi-source information fusion according to claim 1 is characterized in that: The vibration-ultrasound fusion sensor is a compression piezoelectric acceleration sensor.

5. The GIS partial discharge fault detection device based on DS evidence multi-source information fusion according to claim 1 is characterized in that: The optical sensor uses the Jingbang Technology TP3050_SIPM silicon photomultiplier tube as the photoelectric conversion unit of the fluorescent fiber optic detection method.

6. The GIS partial discharge fault detection device based on DS evidence multi-source information fusion according to claim 1 is characterized in that: The coordinator is used to build a ZigBee network, and the coordinator sends data to the background terminal through the GPRS network.

7. A GIS partial discharge fault intelligent detection method based on DS evidence multi-source information fusion, using the GIS partial discharge fault detection device based on DS evidence multi-source information fusion as claimed in claim 1, characterized in that: include: The signal acquisition module collects electromagnetic wave signals, mechanical vibration signals, ultrasonic signals and partial discharge light radiation signals generated by partial discharge inside GIS equipment; The GIS partial discharge fault detection module processes, stores and sends the signals collected by the signal acquisition module; The backend terminal receives the processing information from the GIS partial discharge fault detection module and determines the fault type of the GIS equipment.

8. The GIS partial discharge fault intelligent detection method based on DS evidence multi-source information fusion according to claim 7 is characterized by: The GIS partial discharge fault detection module processes, stores and sends the signals collected by the signal acquisition module; specifically includes: The mechanical vibration signal is processed by wavelet packet transform to obtain the vibration time-frequency analysis feature quantity; The UHF signal, ultrasonic signal, and optical signal spectrum are extracted through CNN to obtain ultra-high frequency feature quantities, acoustic feature quantities, and optical feature quantities; The vibration time-frequency analysis feature quantity, ultra-high frequency feature quantity, acoustic feature quantity and optical feature quantity are normalized and converted into vector machine input data.

9. The GIS partial discharge fault intelligent detection method based on DS evidence multi-source information fusion according to claim 8 is characterized by: The backend terminal receives the processing information of the GIS partial discharge fault detection module and determines the fault type of the GIS equipment, specifically including: Processing the vector machine input data through the support vector machine to obtain the diagnosis result; DS decision fusion is used to determine the type of GIS partial discharge fault based on the diagnosis results.

10. The GIS partial discharge fault intelligent detection method based on DS evidence multi-source information fusion according to claim 9 is characterized by: The GIS partial discharge fault types include: normal state, contamination, air gap, suspended electrode, protrusion, mechanical mild vibration, mechanical moderate vibration and mechanical severe vibration.

Images