基于支持向量机和模态频率的结构损伤诊断

STRUCTURE DAMAGE MONITORING BASED ON SUPPORT VECTOR MACHINE AND MODAL FREQUENCY

支持向量机(supportvectormachine,SVM)是一种基于统计学习理论的机器学习算法,能够较好地解决小样本的学习问题。文中介绍支持向量机回归算法,并应用于结构损伤诊断领域;构造基于模态频率的损伤标识量,作为特征参数训练支持向量机实现对结构损伤的定位和程度标识;最后以梁的损伤识别为例进行验证。结果表明,支持向量机在结构损伤诊断领域中具有很好的应用前景。

Support vector machine（SVM） is a machine learning algorithm based on statistical learning theory, which can solve small-sample learning problems better. The SVM regression algorithm is introduced and is applied to the structure damage monitoring. Damage features formed by vibration modal parameters are used as characteristic parameters to train the SVM to realize the location and .severities identification of structure damage. The results of the rectangular beam＇ s damage identification prove that the SVM is a powerful and promising method for damage monitoring.