留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于动态集成算法的航空发动机气路参数预测

张一震 钟诗胜 付旭云 林琳

张一震, 钟诗胜, 付旭云, 林琳. 基于动态集成算法的航空发动机气路参数预测[J]. 航空动力学报, 2018, 33(9): 2285-2295. doi: 10.13224/j.cnki.jasp.2018.09.027
引用本文: 张一震, 钟诗胜, 付旭云, 林琳. 基于动态集成算法的航空发动机气路参数预测[J]. 航空动力学报, 2018, 33(9): 2285-2295. doi: 10.13224/j.cnki.jasp.2018.09.027
Aeroengine gas path parameter prediction based on dynamic ensemble algorithm[J]. Journal of Aerospace Power, 2018, 33(9): 2285-2295. doi: 10.13224/j.cnki.jasp.2018.09.027
Citation: Aeroengine gas path parameter prediction based on dynamic ensemble algorithm[J]. Journal of Aerospace Power, 2018, 33(9): 2285-2295. doi: 10.13224/j.cnki.jasp.2018.09.027

基于动态集成算法的航空发动机气路参数预测

doi: 10.13224/j.cnki.jasp.2018.09.027
基金项目: 国家自然科学基金重点项目(U1533202); 山东省自主创新及成果转化专项(2014CGZH1101);民航科技创新引导资金项目(MHRD20150104)

Aeroengine gas path parameter prediction based on dynamic ensemble algorithm

  • 摘要: 针对单一学习机对航空发动机气路参数预测困难的问题,提出了基于动态加权核密度估计(DWKDE)组合方法的集成预测算法,该组合方法选择测试样本的近邻样本,通过评估学习机在近邻样本的局部性能动态确定各学习机的权值,并基于该权值利用加权核密度估计实现数据序列的集成预测。该组合方法不易受离群值和样本不对称分布的影响,将该组合方法用于AdaBoost.RT和AdaBoost.R2算法,获得了改进后的集成学习算法。实验证明:相比于神经网络和原始集成学习算法,改进后的集成学习算法较好地提高了航空发动机气路参数序列的预测精度,方均根误差(RMSE)指标至少可降低27%。

     

  • [1] 吴瑞.航空发动机状态预测与健康管理中的气路数据挖掘方法研究[D].四川 广汉:中国民用航空飞行学院,2015.WU Rui.Research on gas path data mining method in aeroengine[D].Guanghan Sichuan:Civil Aviation Flight University of China.(in Chinese)
    [2] 钟诗胜,雷达,丁刚.卷积和离散过程神经网络及其在航空发动机排气温度预测中的应用[J].航空学报,2012,33(3):438-445.ZHONG Shisheng,LEI Da,DING Gang.Convolution sum discrete process neural network and its application in areoengine exhausted gas temperature prediction[J].Acta Areonautica et Astronautica Sinica,2012,33(3):438-445.(in Chinese)
    [3] 范庚,马登武.基于组合优化相关向量机的航空发动机性能参数概率预测方法[J].航空学报,2013,34(9):2110-2121.FAN Geng,MA Dengwu.Probabilistic prediction method for aerogengine performance parameters based on combined optimum relevance vector machine[J].Acta Areonautica et Astronautica Sinica,2013,34(9):2110-2121.(in Chinese)
    [4] LIU K,GEBRAEEL N Z,SHI J.A data-level fusion model for developing composite health indices for degradation modeling and prognostic analysis[J].Transactions on Automation Science and Engineering,2013,10(3):652-664.
    [5] 雷达.基于智能学习模型的民航发动机健康状态预测研究[D].哈尔滨:哈尔滨工业大学,2013.LEI Da.Civil aircraft engine health condition prediction based on intelligent learning models[D].Harbin:Harbin Institute of Technology,2013.(in Chinese)
    [6] 张一震.基于集成学习机的航空发动机气路参数预测方法研究[D].哈尔滨:哈尔滨工业大学,2017.ZHANG Yizhen.Study on aeroengine gas parameters prediction based on ensemble learning[D].Harbin:Harbin Institute of Technology,2017.(in Chinese)
    [7] 张春霞.集成学习中有关算法的研究[D].西安:西安交通大学,2010.ZHANG Chunxia.Research on some algorithms in ensemble learning[D].Xian:Xian Jiaotong University,2010.(in Chinese)
    [8] SCHAPIRE R E,FREUND Y,BARTLETT P,et al.Boosting the margin:a new explanation for the effectiveness of voting methods[J].The Annals of Statistics,1998,26(5):1651-1686.
    [9] WU X,KUMAR V,QUINLAN J R,et al.Top 10 algorithms in data mining[J].Knowledge and Information Systems,2008,14(1):1-37.
    [10] LIM C P,GOH W Y.The application of an ensemble of boosted Elman networks to time series prediction:a benchmark study[J].Computational Intelligence,2005,3(2):119-126.
    [11] SUN J,ZUO H,YANG H,et al.Study of ensemble learning-based fusion prognostics[R].Macao:IEEE Prognostics and Health Management Conference,2010.
    [12] HU C,YOUN B D,WANG P,et al.Ensemble of data-driven prognostic algorithms for robust prediction of remaining useful life[J].Reliability Engineering and System Safety,2012,103(3):120-135.
    [13] KOURENTZES N,BARROW D K,CRONE S F.Neural network ensemble operators for time series forecasting[J].Expert Systems with Applications,2014,41(9):4235-4244.
    [14] SOLOMATINE D P,SHRESTHA D L.AdaBoost.RT:a boosting algorithm for regression problems[R].Budapest:IEEE International Joint Conference on Neural Networks,2004.
    [15] 周志华.机器学习[M].北京:清华大学出版社,2016.
    [16] XUE F,SUBBU R,BONISSONE P.Locally weighted fusion of multiple predictive models[R].Vancouver:IEEE International Joint Conference on Neural Networks,2006.
    [17] 王姚瑶.基于非参数核密度估计的沪深 300 股指期货收益率研究[D].南京:南京大学,2015.WANG Yaoyao.Research on yield rate of HS300 index futures based on non-parametric kernel density estimation[D].Nanjing:Nanjing University,2015.(in Chinese)
    [18] SILVERMAN B W.Density estimation for statistics and data analysis[M].London:Chapman and Hall/CRC (Chemical Rubber Company) Press,1986.
    [19] 孙建乐.基于时间序列相似性的股价趋势预测研究[D].重庆:重庆交通大学,2014.SUN Jianle.Stock price trend prediction research based on time series similarity[D].Chongqing:Chongqing Jiaotong University,2014.(in Chinese)
    [20] HYNDMAN R J,KOEHLER A B.Another look at measures of forecast accuracy[J].International Journal of Forecasting,2006,22(4):679-688.
    [21] LI C,HU J W.A new ARIMA-based neuro-fuzzy approach and swarm intelligence for time series forecasting[J].Engineering Applications of Artificial Intelligence,2012,25(2):295-308.
    [22] 李业波,李秋红,黄向华,等.航空发动机气路部件故障融合诊断方法研究[J].航空学报,2014,35(6):1612-1622.LI Yebo,LI Qiuhong,HUANG Xianghua,et al.Research on gas fault fusion diagnosis of aero-engine component[J].Acta Aeronautica et Astronautica Sinica,2014,35(6):1612-1622.(in Chinese)
    [23] 徐亮.航空发动机气路性能参数趋势预测研究[D].四川 广汉:中国民用航空飞行学院,2016.XU Liang.Research ontrend prediction of gas path parameters of aeroengine[D].Guanghan Sichuan:Civil Aviation Flight University of China,2016.(in Chinese)
  • 加载中
计量
  • 文章访问数:  610
  • HTML浏览量:  6
  • PDF量:  545
  • 被引次数: 0
出版历程
  • 收稿日期:  2017-04-12
  • 刊出日期:  2018-09-28

目录

    /

    返回文章
    返回