Prediction research and application of a combination model based on FEEMD-AE and feedback extreme learning machine

doi:10.11949/j.issn.0438-1157.20171399

Abstract

Abstract:

To facilitate feature extraction and prediction of nonstable time series in industrial processes, a combination model was proposed from fast ensemble empirical mode decomposition (FEEMD), approximate entropy (AE), and feedback extreme learning machine (FELM). First, the FEEMD method was used to decompose complex non-stable time series data into relatively stable intrinsic model function components and residuals from high to low frequency. Secondly, complexity of these components by FEEMD decomposition was reduced by AE complexity degree calculation and feature reconstruction. Thirdly, a feedback mechanism based on traditional ELM structure was introduced to create a feedback layer between output layer and hidden layer for memorizing output data of the hidden layer, calculating trending change rate of output data, and dynamically updating output of the feedback layer, such that a feedback extreme learning machine (FELM) was formed to predict the next timepoint output for nonlinear dynamic system. Finally, the combination model was used to simulate purified terephthalic acid (PTA) solvent system with UCI standard data set. The simulation results show that the proposed method can obtain high prediction accuracy, which will provide guidance for operation optimization of actual production processes.

Key words: ensemble empirical mode decomposition, approximate entropy, extreme learning machine, purified terephthalic acid

摘要：

针对现有工业过程非平稳时间序列中的特征提取及预测问题，提出了基于快速集合经验模态分解（fast ensemble empirical mode decomposition，FEEMD）、近似熵（approximate entropy，AE）和反馈极限学习机（feedback extreme learning machine，FELM）的组合模型。首先，针对复杂非平稳时序数据，采用FEEMD方法将其分解为从高频到低频的相对平稳的本征模态函数分量和余项；其次，为解决经过FEEMD分解出来的分量复杂度问题，运用近似熵（AE）计算分量复杂度并进行特征重构，以降低分量复杂性；然后，基于传统ELM结构，通过引入反馈机制，在输出层与隐含层之间增加反馈层用来记忆隐含层输出数据，并计算数据趋势变化率动态更新反馈层输出，形成反馈极限学习机（FELM），对非线性动态系统的下一时刻输出进行预测；最后，将所提出的组合预测模型通过UCI标准数据集与精对苯二甲酸（PTA）溶剂系统进行建模仿真，仿真结果表明，提出的组合模型预测方法能够得到较高的预测精度，为实际生产操作优化提供了一定的指导。

关键词: 集合经验模态分解, 近似熵, 极限学习机, 精对苯二甲酸

CLC Number:

TP29

XU Yuan, ZHANG Wei, ZHANG Mingqing, HE Yanlin. Prediction research and application of a combination model based on FEEMD-AE and feedback extreme learning machine[J]. CIESC Journal, 2018, 69(3): 1064-1070.

徐圆, 张伟, 张明卿, 贺彦林. 基于FEEMD-AE与反馈极限学习机组合模型预测研究与应用[J]. 化工学报, 2018, 69(3): 1064-1070.

References 30

[1]	HUANG N E, SHEN Z, LONG S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings of the Royal Society A:Mathematical Physical & Engineering Sciences, 1998, 454(1971):903-995.
[2]	KOPSINIS Y, MCLAUGHLIN S. Development of EMD-based denoising methods inspired by wavelet thresholding[J]. IEEE Transactions on Signal Processing, 2009, 57(4):1351-1362.
[3]	YU L, WANG S, LAI K K. Forecasting crude oil price with an EMD-based neural network ensemble learning paradigm[J]. Energy Economics, 2008, 30(5):2623-2635.
[4]	李多, 董海鹰, 杨立霞. 基于EMD与ELM的光伏电站短期功率预测[J]. 可再生能源, 2016, 34(2):173-177. LI D, DONG H Y, YANG L X. The short-term power forecasting of photovoltaic plant based on EMD-ELM[J]. Renewable Energy Resources, 2016, 34(2):173-177.
[5]	YANG J H, DOU W. The wind power forecast model based on improved EMD and SVM[J]. Applied Mechanics and Materials, 2014, 694:150-154.
[6]	TANG L, YU L, WANG S, et al. A novel hybrid ensemble learning paradigm for nuclear energy consumption forecasting[J]. Applied Energy, 2012, 93(5):432-443.
[7]	LIU H, TIAN H Q, LI Y F. Four wind speed multi-step forecasting models using extreme learning machines and signal decomposing algorithms[J]. Energy Conversion and Management, 2015, 100:16-22.
[8]	SUN W. Wind speed forecasting using FEEMD echo state networks with RELM in Hebei, China[J]. Energy Conversion and Management, 2016, 114:197-208.
[9]	RUKHIN A L. Approximate entropy for testing randomness[J]. Journal of Applied Probability, 2000, 37(1):88-100.
[10]	XU L S, WANG K Q, WANG L. Gaussian kernel approximate entropy algorithm for analyzing irregularity of time-series[C]//International Conference on Machine Learning and Cybernetics. IEEE, 2005, 9:5605-5608.
[11]	RICHMAN J S, MOORAMAN J R. Physiological time-series analysis using approximate entropy and sample entropy[J]. American Journal of Physiology Heart & Circulatory Physiology, 2000, 278(6):H2039.
[12]	ZHANG G P. Time series forecasting using a hybrid ARIMA and neural network model[J]. Neurocomputing, 2003, 50(1):159-171.
[13]	VAPNIK V N. The Nature of Statistical Learning Theory[M]. New York:Springer-Verlag, 1995:52-123
[14]	李知艺, 丁剑鹰, 吴迪, 等. 电力负荷区间预测的集成极限学习机方法[J]. 华北电力大学学报, 2014, 41(2):78-88. LI Z Y, DING J Y, WU D, et al. An ensemble model of the extreme learning machine for load interval prediction[J]. Journal of North China Electric Power University, 2014, 41(2):78-88.
[15]	徐圆, 黄兵明, 贺彦林. 基于改进ELM的递归最小二乘时序差分强化学习算法及应用[J].化工学报, 2017, 68(3):916-924. XU Y, HUANG B M, HE Y L. Recursive least-squares TD learning algorithm based on improved extreme learning machine[J]. CIESC Journal, 2017, 68(3):916-924.
[16]	彭荻, 贺彦林, 徐圆, 等. 基于数据特征提取的AANN-ELM研究及化工应用[J]. 化工学报, 2012, 63(9):2920-2925. PENG D, HE Y L, XU Y, et al. Research and chemical application of data feature extraction based AANN-ELM neural network[J]. CIESC Journal, 2012, 63(9):2920-2925.
[17]	高慧慧, 贺彦林, 彭荻, 等. 基于数据属性划分的递阶ELM研究及化工应用[J]. 化工学报, 2013, 64(12):4348-4353. GAO H H, HE Y L, PENG D, et al. Research and chemical application of data attributes decomposition based hierarchical ELM neural network[J]. CIESC Journal, 2013, 64 (12):4348-4353.
[18]	LU C J, KAO L J. A clustering-based sales forecasting scheme by using extreme learning machine and ensembling linkage methods with applications to computer server[J]. Engineering Applications of Artificial Intelligence, 2016, 55:231-238.
[19]	HAN M, ZHANG R, XU M. Multivariate chaotic time series prediction based on ELM-PLSR and hybrid variable selection algorithm[J]. Neural Processing Letters, 2017, (3):1-13.
[20]	徐圆, 叶亮亮, 朱群雄. 基于动态记忆反馈的改进ELM故障预测方法应用研究[J]. 控制与决策, 2015, 30(4):623-629. XU Y, YE L L, ZHU Q X. Improved dynamic recurrent-based ELM neural network for fault prediction[J]. Control and Decision, 2015, 30(4):623-629.
[21]	RAFIEI M, NIKNAM T, KHOOBAN M H. Probabilistic forecasting of hourly electricity price by generalization of ELM for usage in improved wavelet neural network[J]. IEEE Transactions on Industrial Informatics, 2016, 99:71-79.
[22]	WU Z, HUANG N E. Ensemble empirical mode decomposition:a noise-assisted data analysis method[J]. Advances in Adaptive Data Analysis, 2009, 1(1):1-41.
[23]	WANG Y H, YEH C H, YOUNG H W V, et al. On the computational complexity of the empirical mode decomposition algorithm[J]. Physica A Statistical Mechanics & Its Applications, 2014, 400(2):159-167.
[24]	LIU H, TIAN H Q, LI Y F. Comparison of new hybrid FEEMD-MLP, FEEMD-ANFIS, wavelet packet-MLP and wavelet packet-ANFIS for wind speed predictions[J]. Energy Conversion & Management, 2015, 89:1-11.
[25]	AKROM N, ISMAIL Z. FEEMD-DR model for forecasting water consumption[J]. Contemporary Engineering Sciences, 2017, 10:273-284.
[26]	TÜFEKCI P. Prediction of full load electrical power output of a base load operated combined cycle power plant using machine learning methods[J]. International Journal of Electrical Power & Energy Systems, 2014, 60:126-140.
[27]	HEYSEM K, PINAR T, SADIK F G. Local and global learning methods for predicting power of a combined gas & steam turbine[C]//Proceedings of the International Conference on Emerging Trends in Computer and Electronics Engineering. ICETCEE, 2012:13-18.
[28]	HE Y L, GENG Z Q, ZHU Q X. Soft sensor development for the key variables of complex chemical processes using a novel robust bagging nonlinear model integrating improved extreme learning machine with partial least square[J]. Chemometrics & Intelligent Laboratory Systems, 2016, 151:78-88.
[29]	HIBARA K, OBARA M, HAYASHIDA E, et al. MPC with on-line disturbance model estimation and its application to PTA solvent dehydration tower[J]. Journal of Chemical Industry & Engineering, 2008, 59(7):1657-1664.
[30]	ZHU B, CHEN Z S, HE Y L, et al. A novel nonlinear functional expansion based PLS (FEPLS) and its soft sensor application[J]. Chemometrics & Intelligent Laboratory Systems, 2017, 161:108-117.