电力系统保护与控制2011，Vol.39Issue(19)：83-88,6.
基于参数优化的最小二乘支持向量机状态估计方法
State estimate based on parameter-optimized least square support vector machines
陈刚 1闫飞 1龚啸 1王烨2
作者信息
- 1. 重庆大学电气工程学院,重庆400044
- 2. 河南省登封市电业局,河南登封452470
- 折叠
摘要
Abstract
In consideration of that parameter selection of least square support vector machines exerts a major influence on state estimate, this paper presents a state estimation algorithm on the basis of parameter-optimized least square support vector machines. Firstly, in the training process of nonlinear regression estimated model , it adopts two-layer grid search strategy and cross validation to dynamically adjust parameters of the LS-SVM algorithm for better reflecting the complexity of the estimated model, thus increases the estimated accuracy. Secondly, owing to that the estimated model maybe produce big errors, this paper appends a robust method to enhance robustness of the estimated model by means of analyzing characteristic of kernel parameter. Numerical simulation indicates that the proposed method possessses higher estimated accuracy and excellent robustness coping with bad data.关键词
状态估计/最小二乘支持向量机/网格搜索/交叉验证/参数优化Key words
state estimate/ least square support vector machines/ grid search/ cross validation/ parameter optimization分类
信息技术与安全科学引用本文复制引用
陈刚,闫飞,龚啸,王烨..基于参数优化的最小二乘支持向量机状态估计方法[J].电力系统保护与控制,2011,39(19):83-88,6.
