Structural Damage Identification Method Based on IMF Model Energy Feature and BP Neural Network

摘要

This paper brings forward a structural damage identification method that integrates energy characteristics of intrinsic mode functions (IMFs) and BP neural networks. This method begins with decomposing acceleration response of structure into the IMFs by EMD and calculates the model energy feature of signal. The input vectors, which are got by combination of IMF model energy feature from the different test points, can be inputted into BP neural networks so that the neural networks can diagnose the fault. The concrete steps is given in this paper, and the feasibility and effectiveness of this method are tested via the simulation experiments.