Research on modal parameters identification of bridge structure based on adaptive signal de-noising method

For signal de-noising approach based on singular value decomposition (SVD), the method of determining the row number (p) of Hankel matrix and the effective rank (r) both are key problems. In this paper, an adaptive signal de-noising approach which based on genetic algorithm (GA) and SVD was proposed. Choosing signal to noise ratio (SNR) as fitness function, GA was introduced to automatically optimize the parameter of p and r. Then inverse SVD was conducted to achieve the de-noised signal. In order to demonstrate the validity of the approach, two numerical simulation signals with different frequency components are employed. The results show that p can be N/4 or N/3 (N is the length of data), r is twice as the number of dominating frequency. As for measured signal, the complication of the frequency components might be taken into consideration. And in order not to miss the true frequency components when dealing with measured signals, r should be more than twice as the number of dominating frequency, but p can still be N/4 or N/3.