| 一种基于向量的鲁棒ARMA系统辨识器 |
| |
| 引用本文： | 胡学骏,赵慧民,陈广义.一种基于向量的鲁棒ARMA系统辨识器[J].中山大学学报(自然科学版),2005,44(2):39-41,48. |
| |
| 作者姓名： | 胡学骏 赵慧民 陈广义 |
| |
| 作者单位： | 佛山科学技术学院机电学院,广东,佛山,528000 |
| |
| 基金项目： | 广东省自然科学基金，广东省佛山市科技发展专项基金 |
| |
| 摘 要： | 提出一种新的基于向量方法的自回归和运动平均(ARMA)模型系统辨识器,并给出了其参数的统计分析模型.应用结果表明,向量ARMA算法和最小二乘法LS算法相比,在一定条件下,其预测误差精度提高了约1.2 dB;且该系统模型不受分离向量参数的影响.使用非线性函数核,系统将会成为一个鲁棒的非线性辨识过程.
|
| 关 键 词： | ARMA建模 支持向量方法 系统辨识器 时间序列 |
| 文章编号： | 0529-6579(2005)02-0039-04 |
A Robust ARMA System Identification Based on the Vector Method |
| |
| HU Xue-jun,ZHAO Hui-min,CHEN Guang-yi.A Robust ARMA System Identification Based on the Vector Method[J].Acta Scientiarum Naturalium Universitatis Sunyatseni,2005,44(2):39-41,48. |
| |
| Authors: | HU Xue-jun ZHAO Hui-min CHEN Guang-yi |
| |
| Abstract: | To present a new approach to auto-regressive and moving average(ARMA) modeling based on the support vector method,a statistical analysis of the characteristics of the proposed method is carried out.The results show, compared SVM-ARMA with LS,precisions of validation prediction error of the SVM-ARMA improved 1.2 dB than LS in some conditions. Besides, the effect of outliers can be cancelled.With using nonlinear kernels,the system will lead to robust, nonlinear system identification procedures. |
| |
| Keywords: | ARMA modeling support vector method system identification times series |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
