Robust Flow Estimation Algorithm of Multichannel Ultrasonic Flowmeter Based on Random Sampling Least Squares

The multi-path ultrasonic flowmeter is widely used in engineering practice, and the flow algorithm is important for its accuracy. The least-squares estimation method is simple and efficient and has good engineering application value. In practical applications, noises are inevitably introduced to the measurement process because of the flowmeter itself or flow-field interference. The results of classical least squares will deviate from reality because it lacks robustness. In this regard, two flow algorithms of multi-path ultrasonic flowmeter are proposed based on least-squares and random-sampling consensus algorithms, which are widely used in the image field. The two algorithms can resist gross errors effectively by avoiding the interference of external points in the sampling points. To verify the effectiveness of the proposed algorithms, we take the double-bend flow field, which is a typical damaged flow field in engineering, as the research object, and then we compare the four algorithms. It can be seen that the two flow algorithms have higher accuracy and better robustness in the presence of interference noise.