摘要
具有损伤自检测功能的智能复合材料是一个多传感器体系结构,对其传感器进行数目及位置优化,具有重要的实用价值,值得深入研究。采用神经网络建立了复合材料冲击损伤检测方法,运用遗传算法并结合神经网络对复合材料损伤检测的3个传感器布置进行了优化,结果得到了穷举法的验证。该遗传神经网络方法具有一般性,可有效地推广到类似的更多传感器位置优化问题。
The smart composite material possessed of self-detecting damage function is a multi-sensor architecture. It has been of great practical value and is worth to researching further to optimize sensors＇ locations and number for the smart composite material. An effective impact damage detection procedure is established by using a neural network approach. A genetic algorithm combined with neural networks is used to determine the optimum three-sensor positions for damage detection on a composite panel. The result is validated against an exhaustive search. The method presented is generic and can be effectively used in similar more sensors＇ placement problems.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2005年第11期1184-1187,共4页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(90205031)资助项目
关键词
神经网络
遗传算法
损伤检测
传感器位置优化
Neural networks Genetic algorithms Damage detection Optimal sensor placement
作者简介
谢建宏,1971年出生,博士,副教授,研究方向为测控技术及智能结构信息处理与计算机仿真.E-mail：jhxie2003@sohu．com．
