Damage detection in a double-beam system using proper orthogonal decomposition and teaching-learning based algorithm

Abstract

This study applies the inverse approach to damage detection in a doublebeam system. A double-beam system made of two parallel beams is connected through an elastic layer. Degradation of the sti ness of beams element, crack occurrence, and partial destruction of the inner layer are considered as di erent types of damage. The time domain acceleration response of the system is measured, and proper orthogonal decomposition is applied to the collected data in order to derive Proper Orthogonal Values (POVs) and Proper Orthogonal Modes (POMs) of the system. E ect of single damage in di erent locations on the POV has been analyzed, and an objective function is de ned using the dominant POV and POM of each beam separately. In order to increase the robustness of the method against noise, the objective function is enriched by adding statistical property of time domain response. The teaching-learning based optimization algorithm is employed to solve the optimization problem. The e ciency of the proposed method for detecting single and multiple damages in the system is demonstrated with and without noise. Simulation results verify the good accuracy of the proposed method for detecting single and multiple damages of di erent types in the system.

Novel solutions to the (un)damped Helmholtz-Duffing oscillator and its application to plasma physics: Moving boundary method

“…), respectively. Equation (2) without [35][36][37] and with [37][38][39][40] including damping term ( e x 2 ) for…”