基于遗传神经网络的智能复合材料损伤检测传感器位置优化的研究  

Optimal Sensor Placement for Damage Detection in Smart Composite Material Based on Genetic Algorithms and Neural Networks

作　　者：谢建宏[1] 张为公[2] 

机构地区：[1]江西财经大学电子学院,南昌330013 [2]东南大学仪器科学与工程系,南京210096

出　　处：《仪器仪表学报》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

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.

关 键 词：神经网络 遗传算法 损伤检测 传感器位置优化 

分 类 号：TB381[一般工业技术—材料科学与工程] TP391[自动化与计算机技术—计算机应用技术]