|
|
|
| Soft measurement for calorific value of cracking fuel gas based on Online SVM algorithm |
| LI Qi-an, GUO Qiang |
| School of Information and Control Engineering, Liaoning Shihua University, Fushun 113001,China |
|
|
|
Abstract A newly adaptive online support vector regression machine (Online SVM) was proposed to improve the generalization ability of soft sensing model of calorific values of fuel gas in the cracker system that was constructed based on historical data. The approach combined the incremental support vector machine (ISVM) with approximate linear dependence (ALD) condition. New independent samples with ALD condition to update the SVM model were determined by calculating the approximate linear dependence (ALD) value between new samples and modeling samples. The influencing factors of calorific value of fuel gas of cracking furnace were analyzed, and an on-line soft sensing model of calorific values for fuel gas of the cracker system was established using Online SVM algorithm. This model consisted of off-line training module and on-line updating module. The off-line training module was mainly used to produce initially soft sensing model of calorific value based on historical data, and the on-line updating module was used to keep high predictive accuracy for on-line model of calorific value through making off-line training module to learn newly independent samples. A series of comparison simulation experiments were carried out between the proposed method and the conventional SVM and LS-SVM methods using synthetic data, benchmark data and calorific value data of cracking fuel gas. The simulation results show that the proposed method can adapt to new conditions with capability of learning new samples adaptively, and can be used for modeling of soft measurement for calorific values of fuel gas in cracker system with slow time-varying character.
|
|
Published: 28 August 2015
|
|
|
基于在线SVM的裂解炉燃料气热值软测量
针对裂解炉燃料气离线热值模型泛化能力差的问题,提出一种具有自适应能力的在线支持向量机(Online SVM)建模方法.该方法将增量式支持向量机(ISVM)与近似线性依靠(ALD)条件相结合,通过计算新样本与建模样本间的近似线性依靠值,选择满足ALD条件的独立新样本更新SVM模型.分析裂解炉燃料气热值的影响因素,并用Online SVM算法建立裂解炉燃料气热值在线软测量模型.该模型由离线训练模块和在线模型更新模块组成.离线训练模块基于离线数据训练得到初始热值软测量模型,在线更新模块通过使离线模型学习线性独立新样本来保证热值模型的在线预测精度.利用合成数据、Benchmark数据与裂解炉燃料气热值数据,将该方法与传统的支持向量机（SVM）与LS-SVM方法进行对比仿真研究.结果表明：该方法能够适应新的工况,具备自适应学习新样本的能力,可以用于具有慢时变特征的裂解炉燃料气系统热值软测量建模.
|
|
| ［1］ 刘崇明.乙烯装置燃料气系统设计缺陷及改进［J］.乙烯工业,2001,13(4)：33-37．
LIU Chong-ming. Ethylene plant fuel gas system design defects and improvement［J］. Ethylene Industry, 2001, 13(4): 33-37．
［2］ 刘漫丹,杜文莉,钱锋.裂解炉燃料气热值的模糊神经网络软测量［J］. 计算机集成制造系统-CIMS,2003,9 (5)：412-416.
LIU Man-dan, DU Wen-li, QIAN Feng. Soft sensing system of fuzzy-neural network for cracking fuel gas enthalpy［J］. Computer Integrated Manufacturing System, 2003,9(5): 412-416．
［3］ 杨思远,徐佩亮,王振雷.基于小波神经网络的热值软测量建模［J］.石油化工自动化,2011,47 (4)：34-37．
YANG Si-yuan, XU Pei-liang, WANG Zhen-lei. Modeling of soft measurement for calorific values based on wavelet neural network［J］. Automation in Petro-Chemical Industry, 2011, 47(4):34-37．
［4］ THAM M T, MONTAGUE G A, MORRIS A J. A soft-sensors for process estimation and inferential control［J］. Journal of Process Control, 1991, 1(1): 314.
［5］ YANG Y X, CHAI T Y. Soft sensing based on artificial neural network［C］∥ Proceedings of the 1997 American Control Conference. America: ［s.n.］, 1997,1: 674-678．
［6］ ROTEM Y, WACHS A, LEWIN D R. Ethylene compressor monitoring using model-based pca［J］. American Institute of Chemical Engineers Journal,2000, 46(9): 314．
［7］ ZHANG H W, LENNOX B. Integrated condition monitoring and control of fedbatch fermentation processes［J］. Journal of Process Control ,2004, 14(1): 41-50．
［8］ KADLEC P, GRBIC R, GABRYS B. Review of adaptation mechanisms for data-driven soft sensor ［J］. Computers and Chemical Engineering, 2011, 35(1): 124．
［9］ WANG X, KRUGER U, LENNOX B. Recursive partial least squares algorithms for monitoring complex industrial processes ［J］. Control Engineering Practice, 2003, 11(6): 613-632．
［10］ HE X B, YANG Y P. Variable mwpca for adaptive process monitoring［J］. Industrial and Engineering Chemistry Research, 2008, 47(2): 419-427．
［11］ LI W H, YUE H H, VALLE C S. Recursive pca for adaptive process monitoring ［J］. Journal of Process Control, 2000, 10(5): 471-486．
［12］ QIN S J. Recursive pls algorithms for adaptive data modeling［J］. Computers and Chemical Engineering,1998, 22(4):503-514．
［13］ WANG W, CHAI T Y, YU W. Modeling component concentrations of sodium aluminate solution via hammerstein recurrent neural networks ［J］. IEEE Transactions on Control System Technology, 2012, 20(4): 971-982．
［14］ WANG X, KRUGER U, IRWIN G W. Process monitoring approach using fast moving window pca ［J］. Industrial and Engineering Chemistry Research, 2003, 11(6): 613-632.
［15］ CHOI S W, MARTIN E B, MORRIS A J. Adaptive multivariate statistical process control for monitoring time-varying processes［J］. Industrial and Engineering Chemistry Research, 2006, 45(9): 3108-3118．
［16］ FAISAL A, SALMAN N, YEONG K Y. A recursive pls-based soft sensor for prediction of the melt index during grade change operations in HDPE plant ［J］.Korean Journal of Chemical Engineering, 2006, 26(1): 14-20．
［17］ CAUWENBERGHS G, POGGIO T. Incremental and decremental support vector machine learning ［C］∥ Proceedings of the 2001 in Advances in Neural Information Processing Systems. Spain: NIPS, 2001, 13:409-415.
［18］ VLADIMIR N V. Statistical learning theory ［M］. New York: Wiley, 1998.
［19］ LASKOV P, GEHL C, KR UGER S. Incremental support vector learning: analysis, implementation and applications［J］. Journal Machine Learning Research, 2006, 7(1):1909-1936．
［20］ MA J S, THEILER J, PERKINS S. Accurate on-line support vector regression ［J］. Neural Compute,2003,15(11): 2683-2703．
［21］ GIOVANNI M,FRANCESCO P. Learning to trade with incremental support vector regression experts［C］∥ HAIS '08 Proceedings of the 3rd International Workshop on Hybrid Artificial Intelligence Systems. Spain: HAIS, 2008, 5271: 591-598．
［22］ ENGEL Y, MANNOR S, MEIR R. The kernel recursive least squares algorithm［J］. IEEE Transactions on Signal Processing, 2004, 52(8): 2275-2285．
［23］ TANG J, YU W, CHAI T Y. On-line principal component analysis with application to process modeling［J］. Neuro Computing, 2012, 82(1): 167-178．
［24］ 张照娟. 动态模糊神经网络的研究及在燃料气热值软测量中的应用［D］.上海：华东理工大学,2008．
ZHANG Zhao-juan. Research on dynamic fuzzy neural network and its application to soft sensing for calorific value［D］. Shanghai:East China University of Science and Technology,2008．
［25］ 孙优贤,褚健.工业过程控制技术(方法篇)［M］. 北京：化学工业出版社,2005:376-383．
［26］ JEROME H F. Multivariate adaptive regression splines［J］. The Annals of Statistics, 1991, 19(1): 167．
［27］ SUYKENS J A K, BRABANTER J D, LUKAS L.Weighted least squares support vector machines: robustness and sparse approximation［J］. Neurocomputing, 2002, 48(1): 85-105． |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
