Similarity-Based AIC Time-of-Flight Estimation Algorithm for Ultrasonic Flowmeters

Due to the advantages of noninvasive, no pressure drop, bi-direction measurement, and not affected by the liquid conductivity, ultrasonic transient-time flowmeter has been widely used in industrial process. The measurement quality of the ultrasonic flowmeter largely depends on the time-of-flight (TOF) estimation accuracy. However, for the flow rate monitoring in down-hole pipelines of oil production well, the conventional TOF estimation algorithm may lack the noise robustness against waveform distortion, spike noise and interference, and the temporal resolution for flowrate changes of small-diameter pipe. In this article, a similarity-based Akaike information criterion (SAIC) TOF estimation algorithm is proposed. The proposed method uses a similarity-based function to improve the accuracy and noise robustness and uses the subsample TOF corrections to improve temporal resolution. Numerical simulations and experimental studies have been conducted to validate the performance of the proposed method. Results show that the proposed SAIC method significantly reduced the TOF estimation standard deviation from 106.1 to 1.24 ns, and the corresponding flow rate achieved a good accuracy with averaged relative errors of 1.31% and averaged repeatability errors of 0.59%, which make it well suited for industrial applications. The code of the proposed method is open source and can be accessed via the MATLAB file exchange (https://ww2.mathworks.cn/MATLABcentral/fileexchange/169821-similarity-based-aic-time-of-flight-estimation-algorithm ).