Research On Crack Detection Method Of GIS Single-phase Basin Insulator Based On Ultrasonic

The research content of this paper is one of the important sub-topics of the science and technology research project of State Grid Shanxi Electric Power Company "Research and Application of the Key Technology of GIS Basin Insulator Stress Detection Based on Ultrasonic".It is put forward in view of the bad operating environment of GIS Basin Insulator and the frequent crack phenomenon,which may cause malignant accidents.Gas Insulated Switchgear(GIS)is widely used in UHV substations because of its advantages such as high reliability,short installation period and long maintenance period.As the most important insulating component in GIS,the basin insulator plays the role of electrical insulation,gas seal,shell reinforcement and support.In the actual production and charged operation,because of the bad product production process control,flange bolt fasteners stress distribution imbalance and circuit breaker,bus guide bar stress points closing vibration caused by factors such as the basin insulator uneven may make it out of shape,craze,leading to the GIS flat,partial discharge even insulator creepage flashover failure,eventually leading to the occurrence of a large area of power failure.At present,the crack detection methods for basin-type insulators are mainly acoustic emission detection,X-DR imaging detection,laser ultrasonic detection,etc.,but these detection methods are often limited and cannot be put into use in the field.Therefore,it is of great practical significance to study an efficient,accurate and universal method for the crack detection of basin-type insulators to ensure the safe and reliable operation of GIS.The basin insulator is made of composite material,and its geometrical structure is complex.However,the sensitivity of different ultrasonic detection methods may vary greatly due to the different materials and geometrical structures of the workpiece.Therefore,the types and characteristics of ultrasonic wave and the general ultrasonic detection method are studied.The method of crack detection for basin insulators using Lamb wave and two crack evaluation indexes,signal amplitude and energy decay rate,are proposed.The governing equation of Lamb wave in isotropic material is obtained from the stress-strain constitutive equation,motion equation and strain-displacement coordination equation of particle in elastic medium,and the Helmholtz decomposition is carried out.Finally,the dispersion equation characterizing the propagation characteristics of Lamb wave in basin-type insulator is obtained and the corresponding dispersion curve is plotted.Analysis of dispersion curves shows that Lamb waves have dispersion characteristics and multimodal characteristics when propagating in basin insulators.Therefore,it is necessary to select the appropriate signal excitation frequency to achieve the best detection effect when the basin insulator crack detection is carried out.Finite element analysis software is an effective tool for scientific research,which can be used to verify the reliability of theoretical derivation and characteristic analysis.Based on the finite element software COMSOL,the ultrasonic crack detection model of basin-type insulator is built,and the simulation and analysis of the ultrasonic crack detection method of insulator are realized by using the model.The simulation results show that the ultrasonic guided wave has obvious reflection and scattering phenomenon at the crack defect,and there is an obvious acoustic shadow area at the back of the crack.Although the guided wave has diffraction phenomenon at the edges of both sides of the crack,the energy of the diffraction wave has a great degree of attenuation.Further analysis of the simulation data shows that the energy attenuation rate of the signal received by the sensor increases with the increase of the insulator crack,and the closer to the defect location,the greater the change of the energy attenuation rate.This simulation conclusion basically verifies the feasibility of using ultrasonic wave to detect and locate the crack on the basin insulator.Ultrasonic detection signals often contain a lot of noise signals,which will bring some difficulties to signal analysis and defect identification.Therefore,the basic conditions,principles and steps of Empirical Mode Decomposition(EMD)noise reduction and wavelet threshold noise reduction are analyzed,and a wavelet threshold noise reduction method based on EMD and correlation coefficient is proposed.This method can classify each IMF component of ultrasonic detection signal decomposed by EMD through correlation coefficient,and combine with wavelet threshold algorithm to effectively filter out the noise signal in the detection signal and improve the signal-to-noise ratio of the detection signal.The correctness and rationality of the theory and simulation analysis need to be verified by experiments.Therefore,an experimental platform for ultrasonic crack detection of basin-type insulator has been established and a reasonable experimental scheme has been developed.According to actual operation condition,simulating the basin insulator of radial and radial crack,through the sensor receives the signal amplitude and energy of contrast and analysis,found that with the increase of crack size,the received signal amplitude and energy attenuation increases gradually,and the amplitude of received signal in different areas of the sensor and is more sensitive to the crack energy rate for the corresponding area.It is also found that the energy variation of the signal received by the sensor is more sensitive than the amplitude variation when detecting the tiny crack.Therefore,when ultrasonic wave is used for actual detection,the two indicators of amplitude attenuation rate and energy attenuation rate of the detection signal can be comprehensively considered to realize the crack detection and location of the basin insulator and effectively improve the sensitivity of the detection of tiny cracks.