标题：The Hammerstein Predict Model of Control Valve Based on Least Square Support Vector Machine
作者：Huang, Ai-Qin; Wang, Yong
作者机构：[Huang, Ai-Qin; Wang, Yong] Shandong Univ, Sch Mech Engn, Jinan 250100, Peoples R China.
会议名称：4th International Conference on Mechanical Science and Technology (ICMSE 2014)
会议日期：JAN 02-04, 2014
来源：MECHANICAL SCIENCE AND ENGINEERING IV
关键词：Control valve; Hammerstein model; LS-SVM; Prediction
摘要：Control valve is a kind of essential terminal control component controlling the parameters of fluid such as flow and pressure in process-control. However it is a complex nonlinear, multi-input and single-output (MISO) system that is hard to model by traditional methodologies. To establish the pressure model of control valve, this paper presents a Hammerstein modeling method based on the least squares support vector machines (LS-SVM). The linear model parameters and the static nonlinearity of Hammerstein model can be obtained simultaneously by solving a set of linear equations followed by the singular value decomposition (SVD). As an example, a set of actual production data from a controlling system of chlorine in the salt chemistry industry were applied. The simulation results demonstrate that the obtained LS-SVM Hammerstein model can efficiently approximate the pressure of a control valve. Furthermore, the proposed LS-SVM Hammerstein model can be used in artificial intelligent control and the default diagnosis.