Processing ultrasonic signals to identify fluid contents in transit-time flowmeters

In ultrasonic flow measurements, ultrasonic waves are affected by particles and air bubbles as they propagate through the multiphase fluids. Consequently, the propagating waves become attenuated, scattered, distorted, and absorbed. Hence, the signals picked up by the receiver becomes wakened and delayed in time. This weakening and delays in the signals can be compensated by various signal processing techniques to minimise the flow measurement errors. Nevertheless, as shown in this paper the variations in the signals due to multiphase nature of the fluid carry much useful information about the process. The variations in the signals may be used to identify properties and types of materials in the fluid, In this paper, the effects of air bubbles in the fluid are discussed from both theoretical and practical points of view. The analysis of the signals by the application of wavelet theory is presented. It is shown that wavelet application provides useful information on the densities and sizes of air bubbles.