An Approach To Implicit Modelling For Complex Process Optimization

X.G. Yuan,W.Z. An,Y.J. Liu,Y.Q. Luo,C.J. Liu
DOI: https://doi.org/10.1016/S1570-7946(06)80124-0
2006-01-01
Abstract:The difficulties for the complex process optimization come not only from the combinatorial nature and the nonlinearity of the problems, but also from how the problems could be formulated. The complex process is defined in this paper as that a model capturing the behavior of the process is not available, whereas the description of the process could come at lower-scale level. The situation of complex so defined could be encountered in cases for process synthesis and multi-scale modelling. This paper presents a randomized search strategy for tackling this difficulty. Within the proposed search strategy, an explicit formulation including objective function as well as constraint equations is replaced by 'right-away' evaluations of the criteria. The behavior of a randomized search methodology and an example of application of the approach on optimization of thermally coupled complex distillation column system is demonstrated. Keywords: complex process, process modeling, process optimization, Mixed-integer Nonlinear Programming, randomized search.
What problem does this paper attempt to address?