噪声与振动控制2017，Vol.37Issue(6)：180-185,6.DOI:10.3969/j.issn.1006-1355.2017.06.036
基于频率识别纤维增强复合材料弧形板分层损伤
Frequency-based Delamination Assessment of Curved Fiber Reinforced Polymer Panels
摘要
Abstract
A method for assessing the delamination damage in fiber reinforced polymer (FRP) curved panels according to the changes of structural frequencies is proposed. First of all, the finite element models of FRP curved panels with and without delamination damages are constructed respectively in ANSYS. The frequencies of the curved panels with different delaminate damages are computed and compared with those of the healthy curved panels. And the frequency changes between the damaged panels and the healthy panels are obtained. Then, three types of inverse algorithms, namely, Artificial Neural Network (ANN), Genetic algorithm (GA) and Surrogate-assisted optimization (SAO), are used respectively to identify the delamination interface, location and size of the damages. It is found that ANN has the lowest prediction accuracy among the three algorithms;GA has much higher prediction accuracy than that of ANN but it needs the most computer time consuming. To save the computation time, the traditional finite element model of the FRP curved panel in GA is replaced by a surrogate model in SAO. It is found that the time consuming for damage detection is reduced to 1/163 of that of directly using GA, with a slight loss of prediction accuracy.关键词
振动与波/纤维增强复合材料/弧形板/分层损伤识别/遗传算法/人工神经网络Key words
vibration and wave/fiber reinforced polymer/curved panel/delamination damage detection/genetic algorithm/ANN分类
信息技术与安全科学引用本文复制引用
潘静雯,袁浩凡,张芝芳..基于频率识别纤维增强复合材料弧形板分层损伤[J].噪声与振动控制,2017,37(6):180-185,6.基金项目
国家自然科学基金资助项目(51508118) （51508118）
广东省自然科学基金资助项目(2016A030310261) （2016A030310261）
广东省科技计划资助项目(2016B050501004) （2016B050501004）
广州市属高校科技计划资助项目(1201431041) （1201431041）
