Robust Online Process Modeling Method Based on Non-Bias LSSVM
ZHOU Xin-Ran1,2, TENG Zhao-Sheng1, JIANG Xing-Jun3
1.College of Electrical and Information Engineering,Hunan University,Changsha 410082
2.School of Information Science and Engineering,Central South University,Changsha 410075
3.Department of Computer,Hunan Radio and TV University,Changsha 410004
The accuracy of least squares support vector machine(LSSVM) is influenced easily by gross errors and noises superimposed on value measurement of plant output when LSSVM is applied to the dynamic process online modeling directly. Aiming at that problem, robust online process modeling method using non-bias LSSVM is presented after the characteristics of sample sequence structure and noise action are analyzed. During the prediction period, abnormal measure data are recognized and recovered, and measure data containing noises are detected and rectified according to the relation between the predicting error and the set threshold value. Consequently, noises in samples are decreased,and online LSSVM tracks dynamics of process better. The modeling method is robust, and it decreases the effect of gross error and Gaussian white noise on the prediction accuracy of LSSVM to improve the prediction accuracy. The numerical simulation shows the validity and advantage of the proposed method.
[1] Suykens J A K, Vandewalle J. Least Squares Support Vector Machine Classifiers. Neural Processing Letters, 1999, 9(3): 293-300 [2] Suykens J A K. Nonlinear Modeling and Support Vector Machines // Proc of the IEEE Instrumentation and Measurement Technology Conference. Budapest, Hungary, 2001: 287-294 [3] Vapnik V N. Statistical Learning Theory. New York ,USA: Wiley, 1998 [4] Zhang Haoran, Wang Xiaodong. Incremental and Online Learning Algorithm for Regression Least Squares Support Vector Machine. Chinese Journal of Computers, 2006, 29(3): 400-406 (in Chinese) (张浩然,汪晓东.回归最小二乘支持向量机的增量和在线式学习算法.计算机学报, 2006, 29(3): 400-406) [5] Cai Yanning, Hu Changhua. Dynamic Non-Bias LS-SVM Learning Algorithm Based on Cholesky Factorization. Control and Decision, 2008, 23(12): 1363-1367 (in Chinese) (蔡艳宁,胡昌华.一种基于Cholesky 分解的动态无偏LS-SVM 学习算法.控制与决策, 2008, 23(12): 1363-1367) [6] Fan Yugang, Li Ping, Song Zhihuan. Dynamic Weighted Least Squares Support Vector Machines. Control and Decision, 2006, 21(10): 1129-1133 (in Chinese) (范玉刚,李 平,宋执环.动态加权最小二乘支持向量机.控制与决策, 2006, 21(10): 1129-1133) [7] Zhao Yongping, Sun Jianguo. Robust Prediction Model of Least Squares Support Vector Machine Based on Sliding Window. Pattern Recognition and Artificial Intelligence, 2008, 21(1): 1-5 (in Chinese) (赵永平,孙健国.基于滚动窗法最小二乘支持向量机的稳健预测模型.模式识别与人工智能, 2008, 21(1): 1-5) [8] Suykens J A K, de Brabanter J, Lukas L, et al. Weighted Least Squares Support Vector Machines: Robustness and Sparse Approximation. Neurocomputing, 2002, 48(1): 85-105 [9] Wen Wen, Hao Zhifeng, Shao Zhuangfeng, et al. A Heuristic Weight-Setting Algorithm for Robust Weighted Least Squares Support Vector Regression // Proc of the 13th International Conference on Neural Information Processing. Hongkong, China, 2006: 773-781 [10] Suykens J A K, Vandewalle J. Recurrent Least Squares Support Vector Machines. IEEE Trans on Circuits Systems: I, 2000, 47(7): 1109-1114 [11] Song Haiying, Gui Weihua, Yang Chunhua. Application Research of Fuzzy Partial Least Square Support Vector Machine. Journal of System Simulation, 2008, 20(5): 1344-1352 (in Chinese) (宋海鹰,桂卫华,阳春华.模糊偏最小二乘支持向量机的应用研究.系统仿真学报, 2008, 20(5): 1344-1352) [12] Zhang Ying, Su Hongye, Chu Jian. Soft Sensor Modeling Based on Fuzzy Least Squares Support Vector Machines. Control and Decision, 2005, 20(6): 621-624 (in Chinese) (张 英,苏宏业,褚 健.基于模糊最小二乘支持向量机的软测量建模.控制与决策, 2005, 20(6): 621-624) [13] Wu Qing, Liu Sanyang, Du Zhe. Fuzzy Least Square Support Vector Machines for Regression. Journal of Xidian University: Natural Sciences, 2007, 34(5): 773-778 (in Chinese) (吴 青,刘三阳,杜 喆.回归型模糊最小二乘支持向量机.西安电子科技大学学报:自然科学版, 2007, 34(5): 773-778)