Research On Electromagnetic Rail Defect Inspection Data Processing Based On Embedded System

With the rapid development of railway transportation industry in our country, how to ensure traffic safety has been becoming one of the most concerned problems. Among the numerous security technologies, rail defect inspection is the basis significant technology. Rail defect inspection tool named "Railway doctor" is professional vehicle equipment used to detect the rail defect. At present, ultrasonic defect inspection is the major but single inspection method in China. It needs a higher request to the environment when it works. For example, ultrasonic sensors need to contact closely with the rail, and it requires addition of coupling agent. The rail defect inspection cars based on the ultrasonic principle need to be operated manually so that its speed is not fast. At the same time, large-scale cars can only be used under the free railway condition, which is unable to meet the needs of a lot of rail inspection maintenance. This thesis mainly introduces the method of rail defect inspection based on the principle of electromagnetic induction, which calculates the defect information by electromagnetic eddy current inspection of rail feedback. This method can inspect the rail surface and internal defects and has advantages of non-contact, high-speed and so on.This thesis introduces the design of the electromagnetic rail defect inspection data processing platform based on embedded system, which can inspect, store and transmit the information of rail defect. The sine signal applied to the excitation sensor, can generates electromagnetic fields between rail and sensors, and then eddy current that produced on the rail affects the inspection sensor. Demodulating the detection signal can support the rail defect information. The rail defect data and orientation data are stored online, and then transmitted to a remote server. The data processing platform composed of FPGA and embedded ARM division will complete the task together. The main content of the thesis is as follows:This thesis describes the hardware platform design of electromagnetic rail defect inspection data processing based on embedded system. Zedboard development board introduced by Xilinx Company, which is the integration of FPGA and embedded processor ARM, is chosen for the hardware platform. The AD/DA conversion board is used to complete analog to digital and digital to analog conversion work. Electromagnetic rail defect inspection hardware also includes eddy current sensors and amplification circuit for the weak signal, GPS module for the defect location and GPRS module for data transmission.The software platform design of electromagnetic rail defect inspection data processing based on embedded system mainly includes two parts. One part is the program design of FPGA signal demodulation, which uses the FFT algorithm to complete the task, and is programmed by Verilog language. The other part is program design of data management on embedded system, which is programmed by C language, including setting up embedded system, program design of system positioning module, program design of data transmission module, and program design of SQlite embedded database used for storing real-time data.The function of the electromagnetic rail defect inspection data processing platform is tested and verified. Software of serial data acquisition is programmed on PC to receive FPGA signal demodulation data. Curves generated by different frequency acquisition signal are compared to test effectiveness of electromagnetic rail defect inspection. Whether the embedded Linux system and every module can finish their task correctly and timely is tested and verified. The results show that:the data processing platform of the electromagnetic rail defect inspection can realize the established function and meet the requirements of online rail defect inspection.