一种改进的自适应增强-支持向量回归机的故障预测方法

doi:10.3969/j.issn.1000-1093.2012.08.016

兵工学报 ›› 2012, Vol. 33 ›› Issue (8): 991-996.doi: 10.3969/j.issn.1000-1093.2012.08.016

一种改进的自适应增强-支持向量回归机的故障预测方法

邓森₁，景博₁，周宏亮₁，朱海鹏₁，刘小平₂

（1.空军工程大学工程学院, 陕西西安 710038； 2.93050部队，辽宁丹东 118000)

收稿日期:2011-12-08 修回日期:2011-12-08 上线日期:2014-03-04
作者简介:邓森(1986—)，博士研究生
基金资助:
航空科学基金资助项目(20101996012)

Fault Prediction Method Based on Improved AdaBoost-SVR Algorithm

DENG Sen₁, JING Bo₁, ZHOU Hong-liang₁, ZHU Hai-peng₁, LIU Xiao-ping₂

(1.Engineering College, Air Force Engineering University, Xi'an 710038, Shaanxi, China;2.Unit 93050 of PLA, Dandong 118000, Liaoning, China)

Received:2011-12-08 Revised:2011-12-08 Online:2014-03-04

摘要/Abstract

摘要： 针对支持向量回归(SVR)方法对突变故障预测精度较低的问题，提出了一种改进的自适应增强算法(AdaBoost)提升SVR故障预测性能。该方法通过AdaBoost算法获取训练样本中突变点的权重并构造加权支持向量回归机增强突变点的训练，以提高对突变故障预测精度。利用自适应权重裁减方法剔除权重较小的样本点，来提高算法的训练速度。将本文方法用于发动机磨损元素的时间序列预测中，一步预测相对误差达到了0.025. 实验结果表明该方法在保证预测精度的前提下有效地提高了故障预测速度。

关键词: 系统工程方法论, 支持向量回归, 自适应增强算法, 突变故障, 故障预测

Abstract: In order to increase prediction precision of SVR (support vector regression) for catastrophic failures, an improved AdaBoost algorithm was proposed. It could obtain the weights of abnormal data in training sample set by AdaBoost algorithm and a weighted SVR was used to enhance the training of abnormal data, which could improve the prediction precision for catastrophic failure. The samples with small weight were discarded by using an adaptive weight trimming method to improve the training speed. The method was used to predict time series of an engine wear element and the one-step relative prediction error was 0.025. The experiment results demonstrate that the method can improve the speed of fault prediction effectively under desired accuracy.

Key words: methodology of system engineering, support vector regression, AdaBoost algorithm, catastrophic failure, fault prediction

中图分类号:

TP206₊.3

邓森₁，景博₁，周宏亮₁，朱海鹏₁，刘小平₂. 一种改进的自适应增强-支持向量回归机的故障预测方法[J]. 兵工学报, 2012, 33(8): 991-996.

DENG Sen₁, JING Bo₁, ZHOU Hong-liang₁, ZHU Hai-peng₁, LIU Xiao-ping₂. Fault Prediction Method Based on Improved AdaBoost-SVR Algorithm[J]. Acta Armamentarii, 2012, 33(8): 991-996.

参考文献

［1］ Lin G，Peng K，Feng G, et al. Adaboost regression algorithm based on classification type loss[C]∥Proceedings of the 8th Wolrd Congress on Intelligent Control and Automation, Jinan: School of Control Science and Engineering, Shandong University, 2010：978-982.
[2］杨慧中, 邓玉俊. 基于自适应增强算法的支持向量机组合模型[J]. 控制与决策, 2011, 26(2): 316-319.
YANG Hui-zhong, DENG Yu-jun. Compositional model of SVM based on AdaBoosting algorithm[J]. Control and Decision, 2011, 26(2): 316-319. (in Chinese)
[3］牛艳庆, 胡宝清. 基于模糊Adaboost算法的支持向量回归机[J]. 模糊系统与数学, 2006, 20(4): 140-142.
NIU Yan-qing, HU Bao-qing. Support vector machines based on fuzzy Adaboost algorithm[J]. Fuzzy Systems and Mathematics, 2006, 20(4): 140-142. (in Chinese)
[4］蒋焰, 丁晓青. 基于多步校正的改进AdaBoost算法[J]. 清华大学学报:自然科学版, 2008, 48(10): 1613-1616.
JIANG Yan, DING Xiao-qing. AdaBoost algorithm using multi-step correction[J]. Journal of Tsinghua University: Science and Technology, 2008, 48(10): 1613-1616. (in Chinese)
[5］付忠良, 赵向辉, 苗青. AdaBoost算法的推广——一组集成学习算法[J]. 四川大学学报: 工程科学版, 2010, 42(6): 91-98.
FU Zhong-liang, ZHAO Xiang-hui, MIAO Qing. Ensemble learning algorithms: generalization of AdaBoost[J]. Journal of Sichuan University: Engineering Science Edition, 2010, 42(6): 91-98. (in Chinese)
[6］方义秋, 卢道兵, 葛君伟. 一种改进AdaBoost算法的方法[J]. 微电子学与计算机, 2010, 27(7): 40-48.
FANG Yi-qiu, LU Dao-bing, GE Jun-wei. One improved method for adaboost algorithm[J]. Microelectronics & Computer, 2010, 27(7): 40-48. (in Chinese)
[7］ Guo S H, Feng F Z. Study of integrating Adaboost and weight support vector regression model[C]∥International Coference on Artificial Intelligence and Computational Intelligence. Shanghai: Shanghai University of Electric Power, 2009: 258-262.
[8］刁翔, 李奇. 基于加权支持向量机的在线训练算法及应用[J]. 系统仿真学报, 2007, 19(9): 3970-3973.
DIAO Xiang, LI Qi. On-line training algorithm and its application based on weighted SVR[J]. Journal of System Simulation, 2007, 19(9): 3970-3973. (in Chinese)
[9］ Freund Y, Schapire R E. A decision-theoretic generation of online learning and an application to boosting[J]. Journal of Computer and System Science, 1997, 55(1): 119-139.
[10］尉询楷, 李应红, 王硕, 等. 基于支持向量机的航空发动机滑油监控分析[J]. 航空动力学报, 2004, 19(6): 392-397.
WEI Xun-kai, LI Ying-hong, WANG Shuo, et al. Aeroengine lubrication monitoring analysis via support vector machines[J]. Journal of Aerospace Power, 2004, 19(6): 392-397. (in Chinese)

[1]	唐上钦，魏政磊，谢磊，周欢，张卓然. 基于机动单元库的TSO-GRU-Ada机动轨迹预测[J]. 兵工学报, 2022, 43(8): 1913-1925.
[2]	吴一全，宋昱. 基于复无下采样轮廓波和Gaussian小波支持向量回归的红外目标图像背景抑制[J]. 兵工学报, 2015, 36(4): 687-695.
[3]	杨力，马小兵. 维修-更换串联系统贮存可用度建模及费用分析[J]. 兵工学报, 2015, 36(3): 552-558.
[4]	张仲敏，李俊山，宋凭₂，严其飞. 一种基于模糊层次分析的多Agent电子对抗装备状态智能评判方法[J]. 兵工学报, 2014, 35(4): 516-522.
[5]	董华超，宋保维，王鹏. 水下航行器壳体结构多目标优化设计研究[J]. 兵工学报, 2014, 35(3): 392-397.
[6]	曾家有，王国卫，钟建林，谢宇鹏. 多平台舰舰导弹饱和攻击几个协同问题与模型[J]. 兵工学报, 2014, 35(2): 256-261.
[7]	杨丽，孙志礼，印明昂，王宇宁. 基于多属性决策灵敏度分析的供弹机压弹机构优化设计[J]. 兵工学报, 2014, 35(2): 241-247.
[8]	陈欣, 蓝国兴, 段枫,马华东. 基于思维导图的仿真实验方法研究[J]. 兵工学报, 2013, 34(3): 346-352.
[9]	吴勇，潘星，康锐, 马麟. 基于图示评审技术的舰载机航空保障时间分析[J]. 兵工学报, 2013, 34(12): 1611-1615.
[10]	汤扣林. 及时修理系统可靠性建模方法研究[J]. 兵工学报, 2012, 33(8): 980-985.
[11]	杨述明，邱静，刘冠军，杨鹏. 可用度约束条件下装备健康管理信息测试周期优化研究[J]. 兵工学报, 2012, 33(7): 892-896.
[12]	彭博，金星，杨洋，常浩. 基于剩余寿命的潜在故障优化维修方法研究[J]. 兵工学报, 2012, 33(4): 483-487.
[13]	杨述明，邱静，刘冠军，徐玉国，王超. 不确定健康评估下基于状态维修的效果研究[J]. 兵工学报, 2012, 33(2): 252-256.
[14]	赵加鹏，张云海，任翀. 鱼雷外形低阻、低噪和高机动性的多学科优化设计方法[J]. 兵工学报, 2012, 33(2): 186-191.

一种改进的自适应增强-支持向量回归机的故障预测方法

Fault Prediction Method Based on Improved AdaBoost-SVR Algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 14

编辑推荐

Metrics

本文评价