Application of WLS-SVM in Information Fusion for Smart Structures

摘要：

Smart structure is a kind of life-forms multi-sensor architecture, and multi-sensor information fusion is a key technology used to realize the self-diagnosing damages function for smart structures. Due to a few intrinsic flaws of traditional neural networks, Support Vector Machine (SVM) based on Statistical Learning Theory is progressing rapidly in recent years. To overcome the shortcoming as Least Square Support Vector Machine (LS-SVM) lacking sparseness, Weighted Least Square Support Vector Machine (WLS- SVM) used for information fusion is proposed, and its weights are determined by a robust method. Based on the feature-level fusion, WLS-SVM fusion algorithm is applied to self-diagnose damages for smart structures, and compared with LS-SVM under the same conditions. The results show that WLS-SVM obtains the sparseness, and possesses the better dissemination ability and robustness than LS-SVM by introducing the weighted coefficients v<,i>.

关键词： Smart structures Information fusion Weighted least square support vector machine

作者: Jianhong Xie

作者单位: School of Electronics, Jiangxi University of Finance and Economies, Nanchang 330013, China

会议类型: 国际会议

会议名称: 结构、材料与环境健康监测国际会议(International Conference on Health Monitoring og Structure,Material and Environment)

会议地点: 南京

会议语种:英文

页码: 720-724

在线出版日期: 2007-10-16（万方平台首次上网日期，不代表论文的发表时间）