结构自适应序贯正则极端学习机时间序列预测及其应用

推进技术 ›› 2015, Vol. 36 ›› Issue (3): 458-464.

结构自适应序贯正则极端学习机时间序列预测及其应用

何星,王宏力,陆敬辉,姜伟

第二炮兵工程大学控制工程系，陕西西安 710025,第二炮兵工程大学控制工程系，陕西西安 710025,第二炮兵工程大学控制工程系，陕西西安 710025,第二炮兵工程大学控制工程系，陕西西安 710025

发布日期:2021-08-15
作者简介:何星(1985—)，男，博士生，研究领域为智能故障诊断与预测。
基金资助:
国家自然科学基金(61004128)。

Structure Adaptive Sequential Regularized Extreme Learning Machine Based on Time Series Prediction and Its Application

Department of Control Engineering，The Second Artillery Engineering University，Xi’an 710025，China,Department of Control Engineering，The Second Artillery Engineering University，Xi’an 710025，China,Department of Control Engineering，The Second Artillery Engineering University，Xi’an 710025，China and Department of Control Engineering，The Second Artillery Engineering University，Xi’an 710025，China

Published:2021-08-15

摘要/Abstract

摘要： 为提高初始小样本情况下时间序列在线预测的精度，提出了一种结构自适应序贯正则极端学习机(SA-SRELM)。该方法在在线序贯学习阶段，针对不同训练样本规模选择不同的递推方式对输出权值进行更新；同时，在训练样本达到一定规模后，为提高预测模型对系统的动态适应性，在加入新样本的同时对旧样本进行剔除，完成预测模型的训练。利用3种混沌时间序列预测实例对所提方法的有效性进行了验证。最后，将所提方法用于航空发动机排气温度预测中，结果表明该方法相对正则极端学习机(RELM)和序贯正则极端学习机(SRELM)方法具有更好的泛化性能，预测精度分别是二者的约6倍和2倍。

关键词: 正则极端学习机；小样本；时间序列预测；动态适应性；泛化能力

Abstract: To improve the on-line prediction precision of time serials under small initialization sample，a structure adaptive sequential regularized extreme learning machine(SA-SRELM) is proposed. For this new method，at the stage of on-line sequential learning，the output weight of SA-SRELM is renewed by different recursive methods according to training sample size. Meanwhile，when training sample size is added to a destined value，new samples are added and the oldest samples are removed to enhance the dynamic adaptability of prediction model. Based on above process，the prediction model is built. Three typical chaotic time series prediction examples are used to verified the validity of proposed method. In the end，the proposed method is used to predict aeroengine exhaust gas temperature. The results show that SA-SRELM can get better generalization performance compared with regularized extreme learning machine(RELM) and sequential regularized extreme learning machine(SRELM). The prediction accuracy for SA-SRELM is almost six times and twice as high as RELM and SRELM，respectively.

Key words: Regularized extreme learning machine；Small sample；Time series prediction；Dynamic adaptability；Generalization performance

何星,王宏力,陆敬辉,姜伟. 结构自适应序贯正则极端学习机时间序列预测及其应用[J]. 推进技术, 2015, 36(3): 458-464.

HE Xing,WANG Hong-li,LU Jing-hui,JIANG Wei. Structure Adaptive Sequential Regularized Extreme Learning Machine Based on Time Series Prediction and Its Application[J]. Journal of Propulsion Technology, 2015, 36(3): 458-464.

参考文献

[1] Vingerhoeds R A, J anssens P, Netten B D, et al. Enhancing Off-Line and On-Line Condition Monitoring and Fault Diagnosis[J]. Control Engineering Practice,1995, 3 (11): 1515-1528.
[2] Ding G, Zhong S S. Aircraft Engine Lubricating Oil Monitoring by Process Neural Network[J]. Neural Network World, 2006, 16(1): 15-24.
[3] 陈果. 用结构自适应神经网络预测航空发动机性能趋[J]. 航空学报, 2007, 28(3): 535-539.
[4] 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.
[5] 许喆平, 郎荣玲, 邓小乐. 飞机性能参数预测的不确定性处理[J]. 航空学报, 2012, 33(6): 1100-1107.
[6] 于广滨, 丁刚, 姚威, 等. 基于支持过程向量机的航空发动机排气温度预测[J]. 电机与控制学报, 2013, 17(8): 30-36.
[7] 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.
[8] Huang G B, Zhu Q Y, Siew C K. Extreme Learning Machine: Theory and Applications[J]. Neurocomputing, 2006, 70: 489-501.
[9] Yu Q, Miche Y, Eirola E, et al. Regularized Extreme Learning Machine for Regression with Missing Data[J]. Neurocomputing, 2013, 102: 45-51.
[10] Shrivastava N A, Panigrahi B K. A Hybrid Wavelet-ELM Based Short Term Price Forecasting for Electricity Markets[J]. Electrical Power and Energy Systems, 2014, 55: 41-50.
[11] Huang G B, Ding X J, Zhou H M. Optimization Method Based Extreme Learning Machine for Classification[J]. Neurocomputing, 2010, 74: 155-163.
[12] Huang G B, Zhou H M, Ding X J, et al. Extreme Learning Machine for Regression and Multiclass Classification[J]. IEEE Trans on Systems, Man, and Cybernetics-part B: Cybernetics, 2012, 42(2): 513-529.
[13] 邓万宇, 郑庆华, 陈琳, 等. 神经网络极速学习方法研究[J]. 计算机学报, 2010, 3(2): 279-287.
[14] 张弦, 王宏力. 基于贯序正则极端学习机的时间序列预测及其应用[J]. 航空学报, 2011, 32(7): 1302-1308.
[15] 张弦, 王宏力. 利用神经元拓展正则极端学习机预测时间序列[J]. 北京航空航天大学学报, 2011, 37(12): 1510-1514.
[16] Liang N Y, Huang G B, Saratchandran P, et al. A Fast and Accurate On-Line Sequential Learning Algorithm for Feedforward Networks[J]. IEEE Trans on Neural Networks, 2006, 17(6): 1411-1423.
[17] 张贤达. 矩阵分析与应用[M]. 北京: 清华大学出版社, 2005.
[18] 栾圣罡. 基于气路参数样本的航空发动机状态监视方法与系统研究[D]. 哈尔滨: 哈尔滨工业大学, 2008.(编辑:张荣莉)　　　　　　　　　　　　　*　收稿日期:2014-03-31；修订日期:2014-05-16。基金项目:国家自然科学基金(61004128)。作者简介:何星(1985—),男,博士生,研究领域为智能故障诊断与预测。E-mail: trees241@163.com