基于ESN的航空发动机状态组合预测方法

郭阳明; 付琳娟; 冉从宝; 马捷中

基于ESN的航空发动机状态组合预测方法

西北工业大学计算机学院, 西安 710072

基金项目: 国家重点基础研究发展计划; 国家自然科学基金(61001023,61101004); 航空科学基金(2010ZD53039); 陕西省自然科学基金(2010JQ8005)

计量
- 文章访问数: 1767
- HTML浏览量: 4
- PDF量: 990
- 被引次数: 0
出版历程
- 收稿日期: 2012-04-20
- 刊出日期: 2013-04-28

ESN-based combination method for aero-engine condition prediction

School of Computer Science, Northwest Polytechnical University, Xi'an 710072, China

摘要

摘要: 基于回声状态网络(ESN)预测模型,结合小波分析和主元分析,提出一种组合预测方法.首先对含噪非线性时间序列进行小波降噪,并重构时间序列产生训练样本,再将训练样本通过主元分析进行降维处理,降维后的时间序列数据则输入ESN模型进行预测分析.对控制飞机动力输出的动压参数非线性时间序列数据进行了仿真对比实验,结果表明:组合预测方法的5步和单步预测速度累计提高了66.97%,预测的平均平方误差、标准均方根误差和归一化绝对误差也均有较大提高.该方法与传统基于ESN的预测模型相比,能有效地提高预测的效率和精度,是一种有效的非线性时间序列预测方法.
- 航空发动机 /
- 状态预测 /
- 回声状态网络 /
- 小波分析 /
- 主元分析(PCA)
Abstract: As a nonlinear time series prediction method,echo state network (ESN) attracts more attention because of its good approximation capability for the nonlinear system.Aiming at the characteristics of nonlinear time series in the aero-engine's condition prediction analysis,such as noise and presenting chaos,a combination method based on ESN was proposed.Firstly,the noise contained in nonlinear time series was reduced by the wavelet analysis.Then the training sample data were yielded via phase space reconstruction of the time series.After reducing the dimension of the training sample data by principal component analysis,all the remaining principal data were sent into the ESN prediction model.An actual dynamic pressure time series of aircraft power was conducted.The experiments compare the proposed method with traditional ESN prediction model on prediction accuracy and time cost.The results show that the prediction accuracy rate of proposed method within 5 steps and one-step are totally raised 66.97%, and the prediction of the mean square error, normalized root mean square error and normalized absolute error are also improved simultaneously. The proposed method is an effective nonlinear time series prediction method in practice.
- aero-engine /
- health prediction /
- echo state network(ESN) /
- wavelet analysis /
- principal components analysis(PCA)

HTML

参考文献(15)

[1]	郭阳明,蔡小斌,张宝珍,等.故障预测与健康状态管理技术综述[J].计算机测量与控制,2008,16(9):1213-1216, 1219. GUO Yangming,CAI Xiaobin,ZHANG Baozhen,et al.Review of prognostics and health management technology[J].Computer Measurement & Control,2008,16(9):1213-1216,1219.(in Chinese)
[2]	姜彩虹,孙志岩,王曦.航空发动机预测健康管理系统设计的关键技术[J].航空动力学报,2009,24(11):2589-2594. JIANG Caihong,SUN Zhiyan,WANG Xi.Critical technologies for aero-engine prognostics and health management systems development[J].Journal of Aerospace Power,2009,24(11):2589-2594.(in Chinese)
[3]	Sorjamaa A,Hao J,Reyhani N,Ji Y,et al.Methodology for long-term prediction of time series[J].Nerurocomputing,2007(70):2861-2869.
[4]	Jaeger H.The echo state approach to analyzing and training recurrent neural networks [R].German:German National Research Center for Information Technology,2001.
[5]	付林娟,翟正军,郭阳明.基于回声状态网络的多变量预测模型的研究[J].计算机测量与控制,2009,17(7):1356-1358,1361. FU Linjuan,ZHAI Zhengjun,GUO Yangming.Research on multivariate prediction model based on echo state networks[J].Computer Measurement & Control,2009,17(7):1356-1358,1361.(in Chinese)
[6]	雷正伟,徐章遂,米东,等.神经网络的预测性能的优化分析[J].计算机测量与控制,2004,12(1):15-16,20. LEI Zhengwei,XU Zhangsui,MI Dong,et al.Optimization analysis of neural network forecast performance[J].Computer Measurement & Control,2004,12(1):15-16,20.(in Chinese)
[7]	陈果.一种转子故障信号的小波自适应降噪新方法[J].振动工程学报,2007,20(3):285-290. CHEN Guo.A new de-noising method for rotor faults signal[J].Journal of Vibration Engineering,2007, 20(3):285-290.(in Chinese)
[8]	游荣义,陈忠,徐慎初,等.基于小波变换的混沌信号相空间重构研究[J].物理学报,2004,53(9):2882-2888. YOU Rongyi,CHEN Zhong,XU Shenchu,et al.Study on phase-space reconstruction of chaotic signal based on wavelet transform[J].Chinese Journal of Physics,2004,53(9):2882-2888.(in Chinese)
[9]	Mallat S,Huang W.Singularity detection and processing with wavelets[J].IEEE Trans on Information Theory,1992,38(2):617-643.
[10]	席剑辉,韩敏. 主成分分析与神经网络的结合在多变量序列预测中的应用[J].控制理论与应用,2007,24(5):719-724. XI Jianhui,HAN Min.Prediction of multivariate time series based on principal component analysis and neural networks[J].Control Theory & Application,2007,24(5):719-724.(in Chinese)
[11]	范明明.基于主成分分析的多变量混沌时间序列预测研究 [D].辽宁大连:大连理工大学,2006. FAN Mingming.Study on predicting multivariate chaotic time series by principle component analysis [D].Dalian Liaoning:Dalian University of Technology,2006.(in Chinese)
[12]	赵文涛,李琼.关于航空发动机健康监测与诊断技术的研究[J].飞机设计,2010,30(6):30-34. ZHAO Wentao,LI Qiong.Research on aero-engine health monitoring and diagnosis methods[J].Aircraft Design,2010,30(6):30-34.(in Chinese)
[13]	宋云雪,彭朋,史永胜.民用发动机状态混沌预测算法[J].航空动力学报,2011,26(3):675-679. SONG Yunxue,PENG Peng,SHI Yongsheng.Chaos arithmetic for civil aero-engine forecasting[J].Journal of Aerospace Power,2011,26(3):675-679.(in Chinese)
[14]	秦奕青,蔡卫东,杨炳儒.非线性时间序列的相空间重构技术研究[J].系统仿真学报,2008,20(11):2969-2973. QIN Yiqing,CAI Weidong,YANG Bingru.Research on phase space reconstruction of nonlinear time series[J].Journal of System Simulation,2008,20(11):2969-2973.(in Chinese)
[15]	付林娟.复杂非线性系统的时间序列预测方法研究 [D].西安:西北工业大学,2009. FU Linjuan.Research on prediction of time series in complex nonlinear system [D].Xi'an:Northwestern Polytechnical University,2009.(in Chinese)