Research On Modal Parameters Identification Of Time-varying Structures Under Ambient Excitation

With the development of sensor technology, wireless communication technology, micro-processing technology, signal acquisition and processing, information fusion and system modeling technology, aircraft operating status and health status monitoring and diagnostic technology based on structural vibration Information has become a research hotspot. Aiming at the characteristics of modal parameters t changing corresponding to structural damage or mechanical failure the system, this paper researches how to use the collection system vibration response information inspired by the work environment to process and analyze the signal. Then the characteristics of the system operation status and fault the modal parameters is identified, providing important information for the aircraft security. In this paper the following research is carried out:(1) This paper studies the second-generation wavelet analysis based on parameter identification method. Through the multi-level wavelet decomposition, sub-band data is received, then the phase information data of each sub-band wavelet ridge line to describe the modal parameters of the original signal is extracted, by the signal of the identification of modal parameters of the model for system State Analysis.(2) The parameter identification transform methods based on EMD and Hilbert is studied. The signal by the EMD decomposed IMF components are all stable. Hilbert transform can get further Hilbert spectrum, the resultant Hilbert spectrum can accurately reflect the physical processes of energy in various frequency scales and time distribution. According to the signal time-varying characteristics of the local adaptive time-frequency decomposition, eliminating the human factor, and has a high time-frequency resolution and good time-frequency clustering. The advantages of EMD method makes it become a powerful new tool for the field of modal analysis of Modal Parameter Identification.(3) This paper studies parameter identification method based on second generation wavelet analysis and Hilbert transform .Hilbert transform method is the earliest signal of time-frequency modal analysis method, only after the second generation wavelet decomposition of the single-scale sub-band signals using Hilbert transform, its definition of instantaneous frequency has a clear physical meaning.(4) In this paper, the method based on EMD and the second generation wavelet parameter identification method combined is discussed. This method has the advantages of both EMD decomposition and the second generation wavelet analysis. First, the vibration signals generated by the singular value decomposition de-noising is processed, overcoming the second-generation wavelet analysis more sensitive to noise problems. Then we apply the EMD decomposition of the correlation coefficient generated in the process to remove the false IMF component, digital filtering to eliminate aliasing mode. In ensuring a high-resolution time-frequency based on the reduction of edge effects, the accuracy of identification results is improved. This method of modal analysis techniques has great significance in evaluating the structure of dynamic performance, prediction and diagnosis of system failures, the monitoring system status and identification systems.The above method has been validated by MATLAB simulation of the stiffness line through the transformation and mutation system and the use of ADAMS software to a mobile particle with a uniform simply supported beam system simulation experiments. The results show that EMD-based decomposition and the second generation wavelet analysis combining Modal test and analysis methods'error is smaller than the second generation wavelet transform-based parameter identification method's, also its edge effect is smaller than the method based on EMD and Hilbert transform than the combination of parameter identification. It has an accurate identification than the second generation wavelet transform with the parameter identification method combined effects .Identification method is more effective than other time-varying systems Modal test and analysis methods.