A frequency domain based technique for condition monitoring of structures using limited randomly measured dynamic responses

Abstract

Condition monitoring of structures is significantly important from the safety and durability consideration. Parameter identification and subsequent damage detection techniques, using inverse static or dynamic approaches, are important tools to improve the mathematical models for monitoring the condition of structure. However, noise in the measured data might lead to unreliable identification of damage in a structural system. The measurement of static and dynamic responses at all degrees of freedom of a structure is also not feasible in practice. Five different numerical models for damage detection of structures are developed for condition monitoring of truss, beam and framed structures in the framework of finite element model with limited static or dynamic responses. The structural properties viz. axial rigidity and bending rigidity are identified at the element level in the updated models of the structural system. Damage at the element level is identified, by comparing the identified structural parameters of the updated model of the system with those of the undamaged state. Proposed numerical models are suitable for practical problem, as it is able to identify the structural parameters with limited static and modal data of first few modes experimentally measured at selected degrees of freedom. Different numerical examples with various damage scenarios are explored to demonstrate the applicability of the proposed models. The models are able to identify the structural damage with greater accuracy from the noisy dynamic responses even if the extent of damage is small. Experimental validation on simple cantilever beams and portal frames, establish the potential of the proposed methods for its practical implementation. Keywords: Condition Monitoring, Damage, Inverse Static and Dynamic Approach, Modal data, Noise