Structural damage identification considering uncertainties based on a Jaya algorithm with a local pattern search strategy and L0.5 sparse regularization

This study develops an improved Jaya algorithm for structural damage identification considering various uncertainties using vibration data. Most studies consider uncertainties, such as measurement noise, modeling uncertainties, and temperature variations separately but few on their coupling effects. On the other hand, the Jaya algorithm may trap local minimums in optimizing complex objective functions. Therefore, a novel modified Jaya algorithm is developed by integrating the one-step K-means clustering, Hooke-Jeeves pattern search, and linear population reduction strategies. The three modification strategies greatly improve the global optimization performance of the standard Jaya algorithm, which is verified by a series of high-dimension test functions. An L-0.5 regularization is applied to improve the ill-posedness of the damage identification problem and ensure the sparsity of the solution. Numerical and experimental studies on two structures show that the proposed algorithm can identify the structural damage accurately, even considering the coupling effects of various types of uncertainties.

Automated summary

This study develops an improved Jaya algorithm for structural damage identification considering various uncertainties. The algorithm integrates clustering, search, and population reduction strategies to enhance the global optimization performance. Numerical and experimental studies verify the accuracy of the proposed algorithm.