Studies of Joint Damage Identification for Frame Structures
In this paper, Improved back-propagation neural networks (shorted for improved BPNN) in which combined parameters are chosen as input vecter is used to identify joint damages of steel frame structures. Forthermore, the noise injection training techniche is introduced to enhance the anti- interference capability of the neural networks. It has proved that the proposed mothed was a good selection to joint damages for frame structures through numeral verification on a four-story frame and experimental verification on a tow-story frame.
Improved BPNN Frame structures Joint damage identification Combined parameters Experimental modal analysis
Hui Li Yongfeng Du Hongli Wang
Institute of Earthquake Protection and Disaster Mitigation, Lanzhou Univ. of Tech., Lanzhou 730050, China
国际会议
南京
英文
419-423
2007-10-16（万方平台首次上网日期，不代表论文的发表时间）