Damage identification of building structure with LM artificial neural networks
LIShouju,LIUYingxi*,WUYuliang,HEXiang,ZHOUYuanpai ian 116024, China )
The inverse problem of the structural damage detection is formulated as an optimization problem, which is then solved by using artificial neural networks. Static measurements of displacements at a few degrees of freedom are used to determine the location and the magnitude of the damaged joints in the structure. Unlike the classical optimum methods, ANN is able to globally converge. To identify the location and the magnitude of the damaged joints using an artificial neural network is feasible and a well trained artificial neural network by LevenbergMarquardt algorithm reveals an extremely fast convergence, a high degree of accuracy and a good robustness.