A two-step method for damage identification and quantification in large trusses via wavelet transform and optimization algorithm

In the present study, a two-step approach for damage prognosis in long trusses is suggested in which the first step deals with locating probable damages by wavelet transform (WT) and static deflection derived from modal data with the intention of declining the subsequent inverse problem variables. Then, in the second step, optimization based model updating method using Artificial Bee Colony (ABC) algorithm will be employed to quantify the predicted damages within an inverse problem. Interestingly, it is shown in the paper that the two-step method greatly aids in declining the number of variables of the model updating process resulting in more precise results and far less computational effort. Besides, the method is found considerably effective especially for damage prognosis of large trusses. In this regard, two numerical examples including noisy data are considered to assess the efficacy of the method for real practical problems. Additionally, the validity of the second step results is investigated using other optimizers namely Invasive Weed Optimization (IWO) and Particle Swarm Optimization (PSO).