Structural damage detection of jointed frame structures using chaotic excitation and attractor analysis

Engineering assembly structures will suffer joint damage inevitably under environmental loads, which may affect the integrity and functionality of structures, even cause the catastrophic event. In the paper, a method using chaotic excitation and attractor geometrical property analysis to identify the joint damage is investigated. This method utilizes the property that different joint conditions will alter the filtered chaotic signal dimension. Damage features are extracted from reconstructed attractors and used to identify the joint condition. A new parameter based on the attractor local variance is tested as damage feature. An experimental jointed frame structures is designed and the joint damage are induced by varying the fastener preload. The results show that this approach works well for detecting the joint damage and the feature is monotonic varying with the damage increase.