Research On Structural Damage Identification Method Based On Multi-scale Chirplet Sparse Signal Decomposition

Damage identification of structures is very important for people's safety.If the structure is damaged,it will cause some changes in the structural behavior,which is reflected in the structural parameter changes.Therefore,structural parameters are often used to identify structural damage,and structural parameters are usually obtained by signal analysis of structural dynamic response data.When the structure is damaged,the generated signals are usually non-stationary signals.However,the methods in time domain and frequency domain are not feasible for the processing of non-stationary signals.So far,there have been many time-frequency analysis methods for non-stationary signals.These methods have their own shortcomings more or less.In recent years,a new non-stationary signal analysis method with good analysis effect,namely multi-scale line frequency modulation basis coefficient signal decomposition method,has emerged.This signal analysis method can decompose the multi-component non-stationary signals whose frequency varies linearly or curvilinearly,and it can also identify the changes of structural parameters well.In this paper,the method of decomposition of multiscale LFM fundamental coefficient signal is used to study and analyze the vibration response signal.Specific research contents are as follows:(1)In this paper,the numerical simulation signals are analyzed and compared by using the multi-scale linear frequency modulation basis coefficient signal decomposition method(MCSSD)and Hilbert-Huang transform method(HHT).(2)In this paper,a numerical model of a single-degree-of-freedom shear frame structure is established to simulate the damage of the structure under different working conditions by changing the stiffness of the structure at a certain time.The calculated acceleration response is decomposed and analyzed to determine the damage identification effect by comparing the frequency changes under different working conditions.(3)In this paper,a numerical model of shear-type frame structure with five degrees of freedom is established.The damage of structure under different working conditions is simulated by changing the stiffness of different floor structures at a certain time.The calculated acceleration response is decomposed and analyzed to determine the damage identification effect by comparing the frequency changes under different working conditions.(4)In this paper,Duffing and Van der Pol nonlinear vibration systems are introduced.The parameters of free vibration and forced vibration systems of the two systems are identified by the method of sparse signal decomposition based on multi-scale linear frequency modulation.The results show that the multi-scale LFM-based sparse signal decomposition method can well identify the system parameters of Duffing and Van der Pol vibration systems.