An Iterative Quadrature Demodulation for High-Speed Rail Inspection

In this paper, an iterative quadrature demodulation algorithm is proposed to process the detected voltage induced in the differential triple coils from online rail inspection at a speed from 30 to 300 km/h. Considering that the average filter has the advantage of simplicity and high suppression in the zero-point frequency, this paper theoretically discusses the moving average filter, the linear accumulative average, and the weighted average filter (WAF) inside the quadrature demodulation algorithm. It is obtained that the WAF as a low-pass filter (LPF) can improve the demodulation's accuracy and reduce computational cost. Also, numerical simulation is carried out to analyze the effects of the weighted average coefficients on demodulation compared with other LPFs. Finally, the experimental platform for high-speed rail inspection is carried out to verify the feasibility of the proposed demodulation algorithm up to 300 km/h. By processing experimental data, it is demonstrated that the iterative quadrature demodulation algorithm has better performance in both accuracy and response at a high speed of over 300 km/h.