Research on ultrasonic thermometry using maximum eigenvalue algorithm based on BP neural network

Ultrasonic thermometry is a kind of emerging acoustic pyrometry technology, and it can break through the limitation of various electromagnetic environment, where the traditional temperature measurement methods have failed to operate satisfactorily. The temperature on the current path can be retrieved by the speed of the ultrasonic wave. When the measurement path is fixed, the key issue of ultrasonic thermometry is the accurate measurement of ultrasonic time-of-flight. This paper proposes a maximum eigenvalue algorithm based on BP neural network, which has high precision and strong anti-interference ability in ultrasonic time-of-flight measurement. A pair of ultrasonic transducers are installed in a single path for transmitting and receiving signal. With and without heating source, the experimental results show that the temperature obtained by proposed algorithm is closer to the actual temperature than the temperature obtained by threshold and cross-correlation, which proves the effectiveness and reliability of this algorithm.