Hardware Based Real-time Processing of Online Ultrasonic Data of Railway Wheel Cracks: Multi-Sampling Rate Algorithm

In the piezoelectric ultrasonic online flaw detection of railway wheels, the transmission efficiency of ultrasonic data is critical to ensure the real-time detection of wheel flaws. Due to the large number of train wheels to be detected, the hardware sampling frequency of the online flaw detection equipment is high, the amount of ultrasonic echo data is large, and the data transmission and analysis processing time is extended. To improve the data processing efficiency, it is necessary to convert the sampling rate of the ultrasound signal to change the original data volume. According to the actual detection requirements, this paper studies the decimation and interpolation algorithm of ultrasonic digital signals and realizes the sampling rate conversion in the Field Programmable Gate Array (FPGA), which reduces data redundancy and improves system transmission and processing efficiency. The effectiveness and reliability of the method are verified in practical applications.