Accurate Measurement of Blast Shock Wave Pressure by Enhanced Sensor System Based on Neural Network.

During blast shock wave pressure measurement, strong mechanical vibrations and shocks can affect the dynamic characteristics of shock wave pressure sensors, introducing measurement errors. To improve measurement accuracy for the compression phase, a specialized buffer device was designed to enhance the sensor's dynamic response to transient pressure rises. Using a double-diaphragm shock tube, the dynamic calibration of the enhanced sensor system was carried out and the influence of the buffer device on the dynamic performance was investigated. A mathematical model based on a backpropagation (BP) neural network was developed to characterize the sensor system, and a dynamic compensation method was implemented to improve the enhanced shock wave pressure sensor system. Experimental results demonstrated that while the buffer device significantly reduced the operational bandwidth of the sensor system, the BP neural network-based dynamic compensation effectively widened the bandwidth and improved measurement accuracy. This research provides a practical solution for high-precision dynamic pressure measurement, specifically targeting the compression phase in complex environments.