PARAMETER-ESTIMATION IN NONLINEAR ALGEBRAIC MODELS VIA GLOBAL OPTIMIZATION

The estimation of parameters in semi-empirical nonlinear models throug h the error-in-variables method has been widely studied from a computa tional standpoint. This method involves the minimization of a quadrati c objective function subject to the model equations being satisfied. D ue to the nonlinear nature of these models, the resulting formulation is nonconvex in nature. The approaches to solve this problem presented so far in the literature, although computationally efficient, only of fer convergence to a local solution df a model which may contain multi ple minima. In this paper a global optimization approach based on a br anch-and-bound framework will be presented to solve the error-in-varia bles formulation. Various estimation problems were solved and will be presented to illustrate the theoretical and computational aspects of t he proposed method. (C) 1998 Elsevier Science Ltd. All rights reserved .