Stochastic gradient parameter estimation of input nonlinear systems using the filtering technique

摘要：

For input nonlinear output error moving average systems with a two-segment piecewise nonlinearity, a data filtering based stochastic gradient algorithm is developed to estimate the parameters of this nonlinear system based on the data filtering. The basic idea is to combine the keyterm separation principle and the data filtering technique, and to decompose the identification model into two models. The simulation results indicate that the proposed algorithm can give more accurate parameter estimates than existing extended stochastic gradient algorithm.

关键词： Hammerstein models key-term separation principle auxiliary model stochastic gradient output error moving average model

作者: Dongqing Wang Feng Ding Shouqing Sun

作者单位: Automation Engineering, Qingdao University, Qingdao 266071, China and with the Lanyan Group Ltd. as School of Internet of Things Engineering, Jiangnan University, Wuxi 214122,China. Qingdao Hismile College, Qingdao 266100, China.

会议类型: 国际会议

会议名称: 2011 International Symposium on Advanced Control of Industrial Processes(2011工业过程先进控制技术国际研讨会)

会议地点: 杭州

会议语种:英文

页码: 374-378

在线出版日期: 2011-05-01（万方平台首次上网日期，不代表论文的发表时间）