NONLINEAR ESTIMATION FOR PARTIAL DIFFERENTIAL EQUATIONS

被引：24

作者：

SEINFELD, JH

机构：

[1] Department of Chemical Engineering, California Institute of Technology, Pasadena, CA

来源：

CHEMICAL ENGINEERING SCIENCE | 1969年 / 24卷 / 01期

关键词：

D O I：

10.1016/0009-2509(69)80010-2

中图分类号：

TQ [化学工业];

学科分类号：

0817 ;

摘要：

The sequential estimation of the states of a process described by a set of nonlinear hyperbolic or parabolic partial differential equations subject to both stochastic input disturbances and measurement errors is considered. A functional partial differential equation of Hamilton-Jacobi type is derived for the minimum least square estimate error, which is solved approximately in the region of the optimal estimate by a second-order expansion. The optimal estimate is given as the solution of an initial value problem. In the linear case the estimator equations represent analogs of the well-known Kalman filter equations for lumped parameter systems. The determination of the state of a process governed by one-dimensional heat equation from noisy measurements is considered. © 1969.

引用

页码：75 / &