Application research of information fusion technology of multi-sensor in level measurement

According to the velocity of sound calibration and the nonlinear rectification in the ultrasonic level meter, a new information fusion algorithm based on neural network is presented. An improved BP algorithm -LM algorithm is used to train the neural network, which can improve the data´s convergent speed. The simulation result shows that the effect of the temperature to the sound velocity would be fused in the level measurement while realizing non-linear compensation of temperature, so the level can be measured accurately.