摘要
针对结构损伤识别问题,提出一种基于多目标优化策略的结构损伤识别智能算法.该算法利用极端学习机为损伤参数指标与每一阶频率建立非线性函数表达式,先将结构的每一阶实际测量频率与函数表达式相减,再把形成的每个表达式作为优化目标,进而得到结构损伤识别的高维多目标优化模型.为提高模型的求解精度,提出了灰色多粒子群协同的多目标优化算法.实验结果表明,该方法能较好地处理结构损伤识别问题.
Aiming at the problem of structural damage identification,we proposed an intelligent algorithm for structural damage identification based on multi-objective optimization strategy. The algorithm used the extreme learning machine as the damage parameter index,and established the nonlinear function expression between damage parameters and every frequency. We first had the structure of the actual structural frequency minus the function expression,and then took the formation of each expression as a optimization objective to obtain a highdimension multi-objective optimization model based on structural damage identification. In order to improve the solution accuracy of the model, we proposed a grey multi particle swarm co-evolutionary algorithm for multi-objective optimization. The experimental results show that these methods can deal with the problem of structural damage identification.
出处
《吉林大学学报（理学版）》
CAS
CSCD
北大核心
2016年第2期303-308,共6页
Journal of Jilin University:Science Edition
基金
吉林省自然科学基金(批准号:201215149)
吉林省教育厅项目(批准号:吉教科合字[2014]第17号
吉教科合[2013]第243号)
关键词
多目标优化
极端学习机
灰色多粒子群协同
结构损伤识别
multi-objective optimization
extreme learning machine
grey multi particle swarms co-evolutionary
structural damage identification
