Study And Application On Damage Identification Of Temperature-sensitive Structure Based On Wavelet Packet Transform

In the process of long-term health monitoring,the change of environmental factors will affect the dynamic parameters of the structure,which will influence the effect of damage identification.In order to avoid the misjudgment of damage identification due to temperature change,this paper extracts the wavelet packet energy spectrum form the structural responses under different temperature through wavelet packet transform,and then the principal component analysis which could eliminate or reduce the effect of temperature on the wavelet packet energy spectrum is used to identification the damage.The wavelet packet transform and principal component analysis are used to identify the damage base on simply supported beam model,ASCE benchmark numerical model and the test data,as well as the ancient long-term monitoring data.The main contents of this thesis are included as following:1)Firstly,the damage index is constructed by wavelet packet transform and the damage identification is studied.A five meters long simply supported beam finite element model is established.The wavelet packet energy spectrum is extracted from the dynamic response of beam finite element model by using the wavelet packet transform.The damage index is calculated to identify the location of damage.At the same time,the comparisons of different damage degree and noise are given to discuss their influence on damage identification results.2)Secondly,the dynamic responses of the simply supported beam finite element model under different temperatures are obtained.The wavelet packet transform and principal component analysis are used to construct the damage index and recognition rate factor to identify the damage of structure under different temperatures,and the effects of damage location,damage degree,and noise on the damage index and recognition rate factor are discussed.The same analysis is used on the Phase I ASCE structural finite element model to show it can identify the damage accurately.3)Thirdly,the environmental excitation data obtained from the Phase II ASCE benchmark model under ambient excitation is analyzed.The damage index to identify the structural damage is constructed through the natural excitation technique and the wavelet packet energy spectrum.4)At last,the measured data is analyzed by the proposed method.Two groups of samples are extracted from the data,and the time interval of them is one year.The free vibration response of the structure is obtained by random decrement technique.Then the wavelet packet transform and the principal component analysis are used to identify the structural damage.