基于模糊神经网络和证据理论的结构损伤识别新方法 - 地震工程与工程 .pdfVIP

28 4 Vol. 28 No. 4 2008 8 JOURNAL OF EARTHQUAKE ENG INEER ING AND ENGINEERING V IBRAT ION Aug. 2008 : 2008) 1 2 姜绍飞 , 张 帅 ( 1. , 3 0002; 2. , 110168) : , , ( FNN), . , , FNN , FNN , DS. , , FNN . , , : ; DS; ; ; : TU317; TU 28. 73 : A A new structural dam age identification m ethod based on fuzzy neural network and eviden tial th eory 1 2 JIANG Shaofei, ZHANG Shuai ( 1. College of CivilEngineering, FuzhouU niversity, Fuzhou 3 0002, China; 2. School of Civil Eng ineering, Shenyang JianzhuUniversity, Shenyang 110168, China) A bstract: In order tom ake fu ll use of mu ltiresource and uncertain data and to mi prove the damage identification accuracy of the objective from a large structural health monitoring system, a construction m ethod of basic belief as signment function and a new datafusion damage identification methodw ere proposed in this paper via constructing an FNN mode.l In th is proposed damage identification method, the original data ispreprocessed and the feature pa rameters are extracted. Thus these parameters are regarded as input vectors of an FNN, this FNN classifier is con structed and then decision resu ltsw ere draw n by th ismodel. Finally, the fusion decisionm aking results are compu ted and drawn by the data fusion algorithm ofDS evidential theory. A 7story shear frame numericalmodelw asuti lized to validate th is proposed method, and a comparison w as made bewt een th ismethod and single FNN classifier. The results show that the