基于支持向量机的结构损伤程度识别研究

Structure damage severity identification based on support vector machine

支持向量机(SupportVector Machine,SVM)是一种基于统计学习理论的机器学习算法,能够较好的解决小样本的学习问题,具有很优秀的回归特性。本文提出了一种基于支持向量机辨识结构损伤程度的方法,构造模态频率作为损伤标识量训练支持向量机对损伤程度进行预测,并以悬臂梁为例进行了仿真计算。结果表明:支持向量机是一种很有潜力的结构损伤识别方法,具有很好的应用前景。

The SVM （Support Vector Machine） is a machine learning algorithm based on statistical learning theory, and it is also a class of regression method with the good generalization ability. The feasibility of applying the SVM to structure damage identification is discussed in this paper. Firstly the SVM regression algorithms are briefly reviewed. Then a method of estimating the severity of the damage using the SVM is investigated： a few modal frequencies are used as inputs to train the SVM, and the damage＇s severity is predicted. Moreover, a cantilever is analyzed using this method. The results show that the SVM is a powerful method for damage identification.