A Hybrid Wavelet Based–Approach and Genetic Algorithm to Detect Damage in Beam-Like Structures without Baseline Data

This paper proposes an optimized damage identification method in beam-like structures using genetic algorithm (GA) without baseline data. For this purpose, a vibration-based damage identification algorithm using a damage indicator called ‘Relative Wavelet Packet Entropy’ (RWPE) was implemented to determine the location and extent of damage. The procedure does not require vibration signals from an undamaged structure because the method of comparing signals from different locations in the existing state was found to be effective. To ameliorate the algorithm, GA was utilized to identify the best choice for “mother wavelet function” and “decomposition level” of the signals by means of the fundamental fitness function to optimize the algorithm. This resulted in the high accuracy of the damage identification algorithm. In addition, this method has eliminated the difficulties in selecting the type of mother wavelet function for damage identification purposes. To investigate the robustness and accuracy of the proposed method, numerical examples and experimental cases with different damage depths were considered and conducted. The results demonstrated that the proposed method has great potential in the identification of damage location and depth of cut in beam-like structures since it does not require the recorded data from an undamaged beam as a baseline for damage detection. Moreover, the relationship between damage location and depth of damage has been evaluated and results showed that the algorithm can be applied to actual structures.