TY - JOUR AB - Many process systems applications comprise large sets of nonlinear model equations, whose participating variables can be split naturally into independent and dependent variable subsets. This structure can be exploited for deterministic global optimization based on a sequential approach, which performs the optimization in the reduced space of independent variables by considering the model as implicit equations. This paper presents a new method for constructing Taylor model estimators of the implicit equation solutions in order to generate tighter lower bounds on the reduced-space optimization problem. The convergence properties of these estimators are analyzed through numerical examples, and the global optimization approach is demonstrated on a numerical case study featuring a discretized PDE system. © 2013 Elsevier B.V. AU - Rajyaguru,J AU - Chachuat,B DO - 10.1016/B978-0-444-63234-0.50163-9 EP - 978 PY - 2013/// SN - 1570-7946 SP - 973 TI - Taylor models in deterministic global optimization for large-scale systems with few degrees of freedom UR - http://dx.doi.org/10.1016/B978-0-444-63234-0.50163-9 VL - 32 ER -