The Design of Online FPGA-based Electromagnetic Tomography System for Rail Crack Inspection

In order to adapt to the efficiency requirements of railway transportation, high speed rail defect inspection method is urgently needed. Because of its advantage of noncontact and noninvasive inspection, Electromagnetic Tomography (EMT) can be an ideal technique to realize online rail inspection. In this system, excitation signal generation, detecting signal demodulation, image reconstruction algorithm and display of imaging results were implemented inside a Xilinx FPGA (Kintex-7), real time images up to 10,000 frames/second can be achieved with image reconstruction algorithm of Linear Back Projection (LBP) implemented by using FPGA logic. In this paper, a type of L shape sensor array was used to meet the railway defect detection requirements, and FPGA was used to generate four individual sinusoidal signals with four different frequencies as simultaneous excitation, thus, switching circuit could be omitted and system implementation could be simplified. To improve the efficiency of demodulation, there were four separated but same-structured demodulation algorithm implemented in FPGA and demodulated 16 amplitude characteristics at each period of calculation, Fast Fourier Transform (FFT) adopted as demodulation algorithm and implemented in FPGA. The results of imaging were shown in Liquid Crystal Display (LCD), which was directly driven by FPGA. Experiment results show that this method is a feasible approach for crack inspection of rail.