ニューラルネットワークと計算力学に基づくシステム同定の検討 : 第3報, 構造物の広範囲欠陥同定法の研究

概要

論文の詳細を見る
As more and more physical based models are being used by the industry not only to improve reliability but also to shorten design cycles. Computer simulation with finite element model (FEM) plays important role in physical modeling technique. For the sake that there are always some differences in physical parameters from those of real structure, which is in a sense dealt as damage. The paper contributes the approach to detect the inconsistency between the real structure and the FEM efficiently, with Learning Vector Quantization (LVQ) neural network using the frequencies and Modal Assurance Criterion. Method for the detection of structure damage in wide range is studied and the preprocess of the training data is proposed to increase the detection accuracy.
一般社団法人日本機械学会の論文
1999-07-25