提出以光纤光栅传感器为传感元件、简支板结构为研究对象、运用粗集理论对结构损伤部位进行检测识别。介绍了用Rough集理论来构建检测对象的知识表示方法,用K-均值量化算法实现连续信息离散化,采用基于Rough集的约简方法来生成目标决策规则。检验样本的实验显示出所提方法获得了满意的检测识别结果,且运算效率远高于神经网络。

Structural Damage Detection Using Rough Set TheoryLI Yan,JIANG De-sheng,LUO Pei,WANG Li-xin(Key Laboratory of Fiber Optic Sensing Technology and Information Processing Ministry of Education,Wuhan University of Technology,Wuhan 430070,China)For the simple supporting plank sensed by fiber bragg grating strain sensing array,an approach on structural damage detection using Rough set theory was researched.Firstly,the expression of the knowledge about the detected objects based on Rough set was presented.The K-mean algorithm was then used to disperse the continuous data.Finally,the rules for objects recognition were achieved by the reduction technology of the decision-making table to realize the structural damage detection.The experiment results demonstrated that the proposed approach had good quality in detection performance and higher operation efficiency.Rough set;fiber bragg grating;damage detection