Identifying the frequency characteristics of the pressure measurement system with a pressure transmission tube using shock tube method

Abstract

The identification of the frequency characteristics of the pressure measurement system with a pressure transmission tube is inevitably affected by the structure size, the transfer model, and complex noise interferences, which limits the achievable accuracy of the dynamic pressure measurements in some special conditions, such as narrow installation space and high temperature environments. This paper proposes a data-driven calibration method for identifying the frequency characteristics of pressure measurement system with a pressure transmission tube by shock tube system. The distorted calibration signal is first corrected to reduce the effect of complex noise by combining the robust local mean decomposition and a main frequency dispersion component cluster scheme. A modified Levenberg-Marquardt algorithm is presented to establish the transfer model of the pressure measurement system based on the corrected calibration signal and the dynamic pressure generated by shock tube. The frequency characteristics of the pressure measurement system is then identified, and the dispersion of the calibration results is quantitatively evaluated through a kernel density estimation assisted Monte Carlo method. A series of calibration experiments for pressure measurement system with pressure transmission tubes are carried out by a shock tube system. Results show that the proposed method is able to reduce the influence of high-frequency noise and improve the calibration results of the multiple frequency characteristics of the pressure measurement system with a pressure transmission tube. Furthermore, the comparative experiments on the dispersion evaluation of calibration results also demonstrate the superiority of the proposed method over the Fourier transformation method in calibration reliability.