GIS equipment ultrahigh frequency partial discharge diagnosis positioning equipment and method

Abstract

The invention discloses an intelligent positioning and detecting system for partial discharge of electrical equipment based on an ultrahigh-frequency partial discharge detection technology, relates to the technical field of electrical equipment detection, and is used for partial discharge detection, positioning and data analysis and processing of gas insulated metal-enclosed switchgear (GIS) and other primary switchgear. Aiming at the problems of incapability of live detection, high cost, complicated operation, poor effect and the like in the existing GIS partial discharge detection technology, the system comprises a hardware device and a software system, wherein hardware is formed by connecting an ultrahigh frequency sensor (4-channel synchronous acquisition, bandwidth of 1.5 GHz and sampling rate of 6.25 GS / s), a digital oscilloscope and a test host through coaxial cables; software is written based on JAVA, partial discharge is identified through a YOLOv5 algorithm, positioning is carried out through a time difference of arrival method, and noise reduction is carried out through an improved VMD algorithm.

Description

Technical Field

[0001] The present invention relates to the field of electrical equipment detection technology, and more specifically to an intelligent positioning detection system for partial discharge of electrical equipment based on ultra-high frequency partial discharge detection technology, which is used for partial discharge detection, positioning, and related data analysis and processing of gas-insulated metal-enclosed switchgear (GIS) and other primary switchgear. Background Art

[0002] As my country's economy enters a stage of high-quality development and people's living standards continue to improve, the entire industry and market are placing higher demands on the quality of power supply. The continued development of the power industry has driven breakthroughs in related technologies, making it crucial to ensure efficient production, maintenance, and testing of power system infrastructure and maintain the normal and stable operation of the power system.

[0003] Gas-insulated metal-enclosed switchgear (GIS) plays a key role in substations, grid connection points, and high-voltage transmission lines. Its primary functions include improving power system reliability, miniaturization, and safety, and reducing electromagnetic interference and noise. These functions provide an effective solution for the high-demand and high-load operation of power equipment. However, insulation failure is the primary type of GIS equipment failure, accounting for 57%, and partial discharge is a key indicator of insulation failure.

[0004] Partial discharge in GIS equipment is primarily caused by poor contact between switching components, the presence of freely moving metal particles, manufacturing defects in insulators, and design and installation. Because the presence, intensity, and location of partial discharge in equipment cannot be directly observed with the naked eye, intelligent localization and detection technologies and systems for partial discharge in GIS equipment have long been a focus of attention and research in the power industry.

[0005] Currently, the main methods for detecting partial discharges in GIS, both domestically and internationally, include optical detection, gas detection, ultrasonic detection, mechanical vibration, and pulse current detection. However, existing detection technologies have numerous drawbacks, including the inability to conduct live detection, which can cause power outages and substantial economic losses during actual detection. The equipment is expensive, increasing testing costs. Operation is complex, requiring high-level expertise from test personnel. Furthermore, these technologies offer limited detection results, making it difficult to accurately detect relevant information related to partial discharges.

[0006] In summary, in view of the problems existing in the existing GIS partial discharge detection technology, it is urgent to study a GIS partial discharge intelligent positioning detection system and method that can realize live detection, has low detection cost, simple operation and good detection effect, so as to meet the power industry's demand for efficient operation guarantee of GIS equipment. Summary of the Invention

[0007] The purpose of the present invention is to provide an intelligent positioning and detection system for partial discharge of electrical equipment based on ultra-high frequency partial discharge detection technology, so as to overcome the many disadvantages of existing partial discharge detection technology for GIS equipment, such as the inability to detect under power, high equipment cost, complex operation and poor detection effect.

[0008] 1. Hardware (1) Partial discharge diagnostic and positioning device for electrical equipment This device is mainly composed of ultra-high frequency sensors, digital storage oscilloscopes, test hosts and other hardware connected by coaxial cables, so as to achieve the real-time detection and positioning function of partial discharge of GIS equipment.

[0009] (2) Sensors Ultra-high frequency (UHF) sensors are used to collect pulse signals generated by partial discharge (PD) within GIS equipment. When PD occurs, the resulting pulse signals rapidly accumulate and rise in frequency within a very short period of time, reaching the ultra-high frequency (UHF) electromagnetic band of several GHz. This sensor boasts powerful data acquisition capabilities, capable of simultaneous acquisition across four channels, with a sampling bandwidth of 1.5 GHz, a sampling rate of 6.25 GS / s, and 12-bit resolution. It not only captures the raw UHF PD waveform but also accurately transmits this signal to an oscilloscope and host computer, providing a rich and accurate data foundation for subsequent analysis.

[0010] 2. Software System (1) Partial discharge diagnosis and location software for electrical equipment This software is written based on the JAVA environment and is closely matched with the detection equipment. It is one of the core functions of the intelligent detection system.

[0011] (2) Partial discharge identification technology Based on the advanced YOLOv5 target detection algorithm, comprehensive and detailed image preprocessing is performed on a large number of UHF raw waveforms, PRPD patterns, and PRPS patterns to generate a training set for in-depth training. When partial discharge occurs in GIS equipment, the sensor displays the corresponding waveform on the oscilloscope, forming PRPD and PRPS patterns. This technology enables the system to accurately identify partial discharge in GIS equipment, providing a reliable basis for timely fault detection.

[0012] (3) Partial discharge signal positioning technology The time difference of arrival method is used to accurately locate the spatial location of multiple partial discharge signals. Because UHF signals require different times to propagate from different locations to different sensors, calculating the time difference between the electromagnetic waves reaching different sensors can accurately determine the location of the partial discharge source relative to the sensors. This further clarifies the specific spatial location of the partial discharge in the GIS equipment, accurately pinpointing the fault location.

[0013] (4) Noise interference elimination technology The device utilizes a modified variational mode decomposition (VMD) algorithm, which automatically separates noise from discharge signals and effectively classifies multi-source partial discharge signals. This significantly reduces false alarms, provides inspectors with accurate information and data, and ensures the reliability of test results.

Claims

1. A GIS equipment ultra-high frequency partial discharge diagnosis and positioning system, characterized in that: include: Hardware device: Consists of a UHF sensor, a digital storage oscilloscope, and a test host connected by a coaxial cable. The UHF sensor is used to collect UHF pulse signals generated by partial discharge inside the GIS equipment. It has a sampling bandwidth of 1.5 GHz, a sampling rate of 6.25 GS / s, and supports 4-channel simultaneous acquisition. Software system: including: The partial discharge recognition module performs feature recognition on PRPD and PRPS maps based on the YOLOv5 target detection algorithm; The positioning module uses the time difference of arrival (TDOA) method to calculate the time delay of the UHF signal reaching different sensors and determine the spatial location of the partial discharge source; The noise processing module separates noise and partial discharge signals through the variational mode decomposition (VMD) algorithm.

2. The system according to claim 1, wherein: The UHF sensor supports real-time acquisition of pulse signals when the GIS equipment is powered on and running.

3. The system according to claim 1, wherein: The positioning module realizes the spatial positioning of the partial discharge source according to the spatial coordinates and signal arrival time differences of multiple sensors.

4. The system according to claim 1, wherein: The noise processing module uses an improved variational mode decomposition algorithm to classify multi-source partial discharge signals.

5. A method for diagnosing and locating partial discharge of GIS equipment based on the system according to any one of claims 1 to 4, characterized in that: The following steps are involved: S1. Use a UHF sensor to collect UHF pulse signals from GIS equipment and generate raw waveform data. S2. Perform time-frequency analysis on the original waveform data to generate PRPD and PRPS spectra; S3. Identify partial discharge features in PRPD and PRPS maps using the YOLOv5 algorithm. S4. If partial discharge is present, calculate the signal delays of different sensors using the time difference of arrival method and determine the location of the partial discharge source based on the sensor coordinates. S5. Use the variational mode decomposition algorithm to separate the signal from the noise and output the PD type and intensity parameters.