Structural damage detection based on the reconstructed phase space for reinforced concrete slab: Experimental study

Abstract

In this paper, a parameter based on geometry changes of the reconstructed multidimensional phase space of the measured vibration signals for structural damage identification is proposed. The choice of the proper delay time steps and embedding dimensions for phase space reconstruction of linear systems is discussed. Using the determined delay time and embedding dimensions, an index Changes of Phase Space Topology (CPST) with multiple embedding dimensions is calculated and then used to identify the structural damage. To demonstrate the reliability of the proposed method, vibration test data corresponding to different damage states of a continuous reinforced concrete slab is used to calculate the CPST value for damage identification. The results indicate that except the measurement points at structural supports, the CPST values at all the measurement points on the structure increase with structural damage level irrespective of the damage location, indicating that using a single or a minimum number of measurement points and their CPST value can effectively identify damage existence in the structure. The traditional modal-based indices are also calculated using the same vibration data for comparison. It is found that the proposed method with CPST is the most sensitive to structural damage than any modal-based index. The results demonstrate that the proposed method with CPST value is very effective in identifying damage existence in the structure. Although it cannot quantify the damage, it can be a good candidate for continuous structural health monitoring because it needs only a few sensors to detect damage existence in the entire structure.