摘要
由于结构整体动态特性对于局部损伤的不敏感性,只有当结构中的损伤严重到显著影响动态特性的程度,才可能造成结构特性参数的明显变化。宏观尺度模型只能识别出构件层次上的损伤,而对于识别结构中的小损伤或者微损伤以及构件连接节点处的损伤则无能为力。建立了高压输电铁塔一致多尺度数值分析模型,对输电铁塔上下曲臂连接节点处构件内部的不同损伤进行了识别。以小波包能量曲率差为损伤因子,利用小波变换技术,建立了高压输电铁塔的智能损伤识别方法,最后利用RBF神经网络对损伤程度给予判定。研究表明,该方法识别效果比较理想。
Dynamic characteristics of the whole structure is insensitive to local damage,so only when damage of the structure seriously affects the dynamic characteristics can local damage make significant changes to structural parameters. Macro-scale model can only identify the injuries at component level,but it is are powerless for the identifications of small injuries or micro-injuries in the structure and micro-damage at component joint. This paper takes the wavelet packet energy difference of curvature as the damage factor to establish an intelligent damage identification method of high-voltage transmission tower through wavelet transform technology. A consistent multi-scale numerical analysis model of high-voltage transmission tower was established,and different damages in the component joint between the upper and lower cranks of the transmission tower were identified,and then the paper gave a judgment on the degree of injury by use of RBF neural network. The study shows that the recognition method proposed in this paper is effective.
出处
《自然灾害学报》
CSCD
北大核心
2014年第6期240-248,共9页
Journal of Natural Disasters
基金
国家自然科学基金项目(51278091
50978049)
吉林省科技计划项目(20120429)
关键词
输电塔
多尺度分析
模态曲率改变率
损伤识别
小波分析
transmission tower
multi-scale analysis
variation ratio of modal curvature
damage identification
wavelet analysis
