Design of an ultrasonic flowmeter using a cow horn-shaped structure and secretary bird optimization algorithm-back propagation neural network algorithm

A novel dual-channel ultrasonic flowmeter based on the time-difference method is proposed, aiming at solving the measurement error due to the installation angle of the transducers and improving the measurement accuracy. The angle error is eliminated by optimizing the ultrasonic propagation path so that it is parallel to the fluid flow direction. The pipeline design is optimized to reduce the pressure loss to ensure high-precision measurements at different flow rates. In addition, in order to solve the measurement accuracy problem caused by the transducer position, the measurement results of the two channels are fused by Secretary Bird Optimization Algorithm-Back Propagation Neural network, which reduces the error of the measurements and improves the overall accuracy of the measurements. The results of system error analysis and uncertainty evaluation show that the calibrated flowmeter has a maximum relative error of 0.6% and a maximum repeatability of 0.7%, which proves its reliability and effectiveness in fluid measurement.