Constrained Modal Decomposition And Identification Method Based On Frequency Domain Response

As the inherent characteristics of the structure,the accurate acquisition of modal parameters and modal analysis can help to improve the operational performance of the structure and determine the health state of the structure.When the components of structural response signal are complex or the modal analysis is difficult,the modal decomposition method based on modern signal processing is also an important method in structural modal parameter identification.In view of the fact that the current modal decomposition methods may have modal aliasing or distortion,a new modal decomposition method for structural modal identification,Constrained modal decomposition,is proposed.The basic idea of the method is that based on the peak characteristic of the single-order modal response in the frequency domain,the target order modal response decomposition is realized by using the linear combination of the structural response,so that it can fully conform to the original vibration characteristics of the structure.It can completely conform to the original vibration characteristics of the structure.Firstly,the characteristics of linear combination of structural response are analyzed.Then,based on the concept of linear combination,a modal decomposition method with peak amplitude constraint of frequency domain response is proposed,and the cyclic optimization of decomposition coefficient and inherent modal parameters is completed by iterative updating.And then,for the case of dense frequency,a modal decomposition method with peak derivative constraint of frequency domain response is developed,and on this basis,in order to make the modal decomposition process more focused,the local peak derivative constraint method is studied,only the frequency components concerned are separated.Then,the accuracy of the constrained modal decomposition method is verified by numerical simulation,and the stability of the algorithm is analyzed,and it is concluded that the amplitude constraint method has stronger noise robustness,while the derivative constraint method has higher modal decomposition accuracy.With regard to the parameter selection of the frequency point of the derivative constraint method,the energy discrimination index is given to quantify the decomposition effect,and the parameter corresponding to the minimum energy value is taken as the best value.Meanwhile,the positive contribution of the signal preprocessing technology of singular spectrum analysis under the influence of noise to signal denoising processing and modal order determination is studied,and its application stability is simply analyzed.Finally,the two span continuous beam numerical model and four story frame structure test model are used to verify the practicability of amplitude constraint method and derivative constraint method for modal decomposition and modal analysis.By comparing with the existing modal decomposition methods,it is proved that this method has higher accuracy of modal decomposition and can achieve the ideal goal of modal decomposition.