Extension of measurable velocity limit of ultrasonic velocity profiling based on phase difference gradient analysis

Abstract A signal processing method, called the ultrasonic cross-spectrum method, is proposed to extend the measurable velocity limit of ultrasonic velocity profiling. It focuses on the phase difference spectrum within the signal bandwidth and extends the velocity limit by analyzing its gradient. The algorithmic performance is confirmed via echo signal simulation and experimental pipe flow measurement. Consequently, setting a threshold for the coefficient of determination for phase-difference spectrum fitting enables us to reduce the false velocity detection rate to <1.6% and achieve an effective data rate of 42% in low signal-to-noise ratio conditions. This data rate is twice that of a conventional method. Moreover, a comparison of pipe flow results between our method and particle image velocimetry shows that the velocity data difference is within ±2% for a velocity range twice the conventional measurable velocity limit. In short, we demonstrate the potential of this method for high-performance velocity profiling.