NOVEL SENSOR PLACEMENT FOR DAMAGE IDENTIFICATION IN A CRACKED COMPLEX STRUCTURE WITH STRUCTURAL VARIABILITY

The focus of this work is on sensor placement for structural dynamic analysis and damage detection. In particular, novel sensor placement techniques are presented for the detection of cracks in ground vehicles and other complex structures. These techniques are designed to provide vibration characteristics for complex structures that have both cracks and structural variability (such as uncertainty in the geometry or the material properties). Such techniques are needed because structural variability affects the mode shapes of a structure, and thus the optimal sensor locations for detecting cracks are affected. Two approaches are developed and used: (a) parametric reduced order models (PROMs), and (b) bilinear mode approximation (BMA). Based on PROMs and BMA, a novel sensor placement method (which uses a derivative of the effective independent distributed vector algorithm) is used to determine the optimal sensor locations for complex structures with cracks and structural variability. The approach can also be used to estimate the crack length. The length is identified by using a few mode shapes and only a few selected measurement locations. The information from the sensors can be used to determine variations in mode shapes of the structure (between healthy and cracked states) for different crack lengths. The variation in mode shapes can then be used to identify the crack length. Numerical results are presented for a ground vehicle frame. The sensor placement method is applied first to find the optimal sensor locations for a structure with a crack and parameter variability, and then to identify the length of a crack.