登录    |    注册

您好,欢迎来到中国测试科技资讯平台!

首页> 《中国测试》期刊 >本期导读>基于AdaBoost-LSSVM的纤维复合材料损伤识别

基于AdaBoost-LSSVM的纤维复合材料损伤识别

234    2020-09-17

¥0.50

全文售价

作者:刘宇韬, 盛文娟

作者单位:上海电力大学自动化工程学院,上海 200090


关键词:AdaBoost算法;玻璃纤维复合材料;最小二乘支持向量机;损伤识别


摘要:

玻璃纤维增强复合材料(GFRP)多种损伤模式之间相互作用难以有效检测, 且获取大量标签数据十分困难。该文选择对小样本分类问题有效且运算速度快的LSSVM作为AdaBoost集成学习框架的弱分类器,提出一种AdaBoost-LSSVM算法,并在UCI数据集上进行有效性验证。在玻璃纤维复合材料损伤识别应用中,首先采用小波包变换提取GFRP在静态压痕试验下释放的声发射信号在各频段下的能量占比作为特征向量,然后采用该算法对纤维分层、界面脱粘、纤维断裂和无损伤4种状态进行识别和分类。实验结果表明,提出的AdaBoost-LSSVM集成学习模型具有较高的识别准确率,相比AdaBoost-DT和单一LSSVM分别提升5.75%和7.5%。其中,该方法在界面脱粘上的识别比AdaBoost-DT和LSSVM分别提升10%和20%,在纤维断裂上均提升9%。


Fiber reinforced polymer damage identification based on AdaBoost-LSSVM
LIU Yutao, SHENG Wenjuan
College of Automation Engineering, Shanghai University of Electric Power, Shanghai 200090, China
Abstract: The failure type of glass fiber reinforced polymer (GFRP) are difficult to detect effectively due to the interaction among multiple damage modes. Moreover, a large amount of labeled damage data is hard to obtain. Because LSSVM is effective for small sample classification problems and has a fast operation speed, it is chosen as the weak classifier of AdaBoost framework. Therefore, AdaBoost-LSSVM algorithm is proposed, and the validity is proved by UCI dataset. In the application of GFRP damage identification, acoustic emission signals are firstly analyzed by wavelet packet transform, the energy ratios in different frequency bands are extracted as feature vectors for damage identification. Then, the proposed algorithm is used to identify and classify four states including delamination, debonding, fiber breakage and non-damage. Experimental results show that AdaBoost-LSSVM has better classification accuracy. The average accuracy of AdaBoost-LSSVM is 5.75% improvement over AdaBoost-DT, and 7.5% improvement over LSSVM. In detail, proposed method is 10% and 20% higher than AdaBoost-DT and LSSVM in identification of debonding, respectively. Also, the accuracy is increased by 9% compared with the other two algorithms in the identification of fiber breakage.
Keywords: AdaBoost algorithm;GFRP;least square support vector machine;damage identification
2020, 46(8):148-153  收稿日期: 2020-03-13;收到修改稿日期: 2020-04-29
基金项目: 国家自然科学基金项目(61905139)
作者简介: 刘宇韬(1997-),男,安徽合肥市人,硕士研究生,专业方向为机器学习、复合材料损伤识别
参考文献
[1] 董慧民, 安学锋, 益小苏, 等. 纤维增强聚合物基复合材料低速冲击研究进展[J]. 材料工程, 2015, 43(5): 89-100
[2] 张丹丹, 孙耀宁, 王雅. 多轴向玻璃纤维增强树脂基复合材料的破坏特性和损伤机制[J]. 复合材料学报, 2017, 34(2): 381-388
[3] 黄展鸿, 黄春芳, 张鉴炜, 等. 声发射技术在纤维增强复合材料损伤检测和破坏过程分析中的应用研究进展[J]. 材料导报, 2018, 32(7): 1122-1128
[4] BRUNNER A J. Identification of damage mechanisms in fiber-reinforced polymer-matrix composites with Acoustic Emission and the challenge of assessing structural integrity and service-life[J]. Construction and Building Materials, 2018, 173: 629-637
[5] GUTKIN R, GREEN C J, VANGRATTANACHAI S, et al. On acoustic emission for failure investigation in CFRP: Pattern recognition and peak frequency analyses[J]. Mechanical Systems and Signal Processing, 2011, 25(4): 1393-1407
[6] 栗丽, 晏雄. 复合材料损伤失效的声发射检测研究进展[J]. 材料导报, 2013, 27(17): 19-22
[7] SAYAR H, AZADI M, GHASEMI-GHALEBAHMAN A, et al. Clustering effect on damage mechanisms in open-hole laminated carbon/epoxy composite under constant tensile loading rate, using acoustic emission[J]. Composite Structures, 2018, 204: 1-11
[8] O'BRIEN R J, FONTANA J M, PONSO N, et al. A pattern recognition system based on acoustic signals for fault detection on composite materials[J]. European Journal of Mechanics-A/Solids, 2017, 64: 1-10
[9] GOMES G F, DE ALMEIDA F A, JUNQUEIRA D M, et al. Optimized damage identification in CFRP plates by reduced mode shapes and GA-ANN methods[J]. Engineering Structures, 2019, 181: 111-123
[10] NAIR A, CAI C S, KONG X. Acoustic emission pattern recognition in CFRP retrofitted RC beams for failure mode identification[J]. Composites Part B: Engineering, 2019, 161: 691-701
[11] 丁文哲, 李新洪, 杨虹. 基于Adaboost的填充式防护结构超高速撞击损伤预测[J]. 北京航空航天大学学报, 2019, 45(1): 149-158
[12] LI L, HUANG Y, TAO J, et al. Featured temporal segmentation method and AdaBoost-BP detector for internal leakage evaluation of a hydraulic cylinder[J]. Measurement, 2018, 130: 279-289
[13] 曹惠玲, 高升, 薛鹏. 基于多分类AdaBoost的航空发动机故障诊断[J]. 北京航空航天大学学报, 2018, 44(9): 1818-1825
[14] 王建国, 杨云中, 秦波, 等. 基于峭度与IMF能量融合特征和LS-SVM的齿轮故障诊断研究[J]. 中国测试, 2016, 42(4): 93-97