基于Kernel-SOM的非线性系统辨识及模型运行收敛性分析
doi: 10.3724/SP.J.1146.2007.00010
Kernel-SOM Based Nonlinear System Identification and Model Running Convergence Analysis
-
摘要: 该文提出基于核SOM(Kernel-SOM)的非监督非线性系统辨识方法。在辨识误差和系统初始输入误差同时存在的条件下,对Kernel-SOM辨识模型独立运行的收敛性进行了理论分析,并给出了辨识模型运行收敛的定理。数字仿真表明了所述方法的有效性及收敛定理的正确性。Abstract: A Kernel-SOM based unsupervised nonlinear system identification algorithm is proposed. Analysis of the model running convergence of the proposed algorithm is performed, and the convergence theorem is proofed by considering both identification error and initial input error. Numerical simulation results demonstrate the effectiveness of the proposed identification algorithm and the correctness of the convergence theorem.
-
Kohonen T. Self-organization map[J].Proc. IEEE.1990, 78(9):1464-1480[2]Barreto G A and Aluizio A F R. Identification and control ofdynamical systems using the self-organizing map[J].IEEETrans. on Neural Networks.2004, 15(5):1244-1259[3]Yu Dong-jun, et al.. Kernel-SOM based visualization offinancial time series forecasting. International Conference oninnovative computing, information and control. Beijing, 2006,Volume II: 470-473.[4]Pan Zhisong, Chen Songcan, and Zhang Daoqiang. AKernel-based SOM classification in input space. ActaElectronica Sinica, 2004, 32(2): 227-231.[5]Scholkopf B, Burges C J C, and Smola A J. Advances inKernel Methods - Support Vector Learning [M]. Cambridge,MA, The MIT Press, 1999: 255-268.[6]Pao Xiaohong, et al.. Model error analysis in nonlinearsystem identification using neural networks (I). Control andDecision, 1997, 12(5): 20-25.[7]Lin C T. Neural Fuzzy Systems. New York: Prentice-HallPress. 1997, Chapter 3.
计量
- 文章访问数: 3321
- HTML全文浏览量: 77
- PDF下载量: 987
- 被引次数: 0