Shock tunnel force measurement signal frequency domain analysis method based on deep learning

Abstract

The invention discloses a shock tunnel force measurement signal frequency domain analysis method based on deep learning. The method comprises the steps of: building a shock tunnel aerodynamic force measurement system, and collecting a plurality of balance sample signals in a time domain based on an SVDC technology; decomposing the balance sample signal by adopting wavelet transform to obtain sub-signals, and performing time-frequency conversion on the sub-signals to obtain effective characteristic signals; performing fast Fourier transform on the effective feature signals in the time domain to obtain frequency domain signals converted to a spectrogram, and performing dimensionless processing on the frequency domain signals; training a convolutional neural network model, and performing intelligent modeling on the frequency domain signals by using the convolutional neural network model to obtain effective output signals after convolution circulation; and performing dimensionalization processing and inverse fast Fourier transform on the effective output signals to obtain filtered aerodynamic force signals in a time domain. Inertial vibration signals are filtered, real aerodynamic force signals are obtained, and reliability and precision indexes of pulse wind tunnel force measurement results are improved.

Description

technical field [0001] The invention relates to the technical field of wind tunnel force measurement signal test, in particular to a frequency domain analysis method for shock wave wind tunnel force measurement signal based on deep learning. Background technique [0002] The measurement method of the pulse strain balance is mostly to measure the voltage change of the strain gauge caused by the deformation of the fast response model under the impact load, and then reflect the model load, because of its large overall structure rigidity, low interference between components, high output sensitivity, The characteristics of strong stability and high precision are widely used in hypersonic vehicle force testing and so on. During the dynamometric test, the dynamometric system generates inertial vibration under the impulse excitation of the instantaneous start of the wind tunnel flow field. Due to the limitation of the effective running time of the wind tunnel and the complexity of t...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a frequency-domain analysis method for shock tunnel force measurement signals based on deep learning, so as to solve the technical problem that the prior art cannot effectively filter out interference signals, thereby affecting the accuracy of aerodynamic measurement

Images