The Signal Processing Base On FPGA Of Electromagnetic Ultrasonic For Rail Nondestructive Testing

With the increase of railway mileage and transportation capacity,the safety monitoring pressure of rail transit in China is increasing day by day.It has led to a rapid increase in the demand for nondestructive testing of rails.At present,the mainstream technology of non-destructive testing of rails is the traditional piezoelectric ultrasonic technology.But its testing speed has been approaching the limit.In recent years,due to the development of Electromagnetic(EMAT)technology,the relative advantage of piezoelectric ultrasonic flaw detection technology has become increasingly prominent.However,EMAT has the disadvantage of low efficiency and weak signal transduction of electromagnetic ultrasonic sensors.In addition,EMAT high-speed NDT require high-speed signal processing to support.To some extent,these problems limit the application of electromagnetic ultrasonic in rail nondestructive testing.Based on the two aspects of electromagnetic ultrasonic sensor and electromagnetic ultrasonic signal processing,the electromagnetic ultrasonic rail NDT system has been comprehensively researched and verified by experiments.The electromagnetic ultrasonic sensor and its signal processing algorithm meeting the application performance requirements have been completed.The contents and results of the specific research are as follows:The dispersion characteristics of Lamb electromagnetic ultrasonic sensor are theoretically calculated by using GUIGUW.And the Comsol Multiphysics finite element analysis software is used to model and simulate the Lamb wave and SV oblique incident electromagnetic ultrasonic sensor,which are relatively less studied at present.Based on the simulation results,the structures of these two types of sensors are studied and optimized.Lamb wave and SV oblique incident electromagnetic ultrasonic sensors are designed.This paper solves the weak electromagnetic signal,low SNR using wavelet denoising algorithm and realizes it by FPGA.After the simulation of the noise reduction process of the electromagnetic ultrasonic signal,it proves that the wavelet threshold denoising algorithm designed in this paper has the highest SNR of 16.2d B compared with the conventional denoising algorithm.Compared with other wavelet algorithm,the implementation method proposed in this paper is easily and meet the requirement of real-time signal process.In this paper,a special test platform has been set up to verify the design of electromagnetic ultrasonic sensor and wavelet de-noising signal processing algorithm.The experimental results show that the sensor and algorithm designed in this paper can meet the performance requirements of non-destructive flaw detection.