Temperature Compensation of Ultrasonic Flow Measurement Based on the Neural Network

In order to eliminate the influence of temperature on ultrasonic flow measurement, on the basis of nonlinear compensation method, temperature compensating model of ultrasonic flow measurement is established by the application of BP (Back Propagation) neural network. The network is adjusted by measuring samples to compensate temperature loss of the ultrasonic flow measurement system. The simulation results show that the influence of temperature on the ultrasonic flow measurement is dramatically reduced to 3% from 5.2% which fulfill the targets of reducing errors and increasing measurement precision. By using neural network technique, the precision of the ultrasonic flow measurement is improved greatly, and the accurate intelligent temperature compensation of the ultrasonic flow measurement is made.