Error modeling of the runway-type magnetohydrodynamic linear vibration sensor based on the SSA-BP neural network

The magnetohydrodynamic (MHD) linear vibration sensor is a new type of vibration measurement sensor, which has the characteristics of fast response and strong anti-interference. In order to improve the measurement accuracy of the MHD linear vibration sensor, the error of the sensor is studied in this paper. Since the singular spectrum analysis algorithm can eliminate the coupling effect of the data and improve the robustness of the error model, this paper combines the singular spectrum analysis algorithm (SSA) with the BP neural network, and use the SSA-BP neural network to establish the error model of the MHD linear vibration sensor. The experimental results show that the average absolute error of the output signal of the MHD linear vibration sensor compensated by the SSA-BP neural network error model was reduced by 1 time, the signal-to-noise ratio was increased by 7 times, and the correlation coefficient was greater than 0.9.