LAUSR

Abstract This paper proposes a general superstructure and a Mixed-Integer Nonlinear Programming (MINLP) model for the synthesis and simultaneous optimisation and Heat Integration (HI) of Single- and Multiple-Effect Evaporation (SEE/MEE) systems including Mechanical Vapour Recompression (MVR) and the background process. The proposed superstructure also includes different flow patterns (forward feed, backward feed, parallel feed and mixed feed), Flashing of Condensates (FCs), single- and multi-stage MVR systems and various HI opportunities for preheating of feed stream (e.g. with condensates, bled vapours, and hot streams from the background process). The newly proposed SEE/MEE-FC-MVR superstructure is combined with a Heat Exchanger Network (HEN) superstructure for performing simultaneous optimisation and HI. On the basis of this combined superstructure, an MINLP model with tight bounds on the variables is developed and implemented for its solution in the General Algebraic Modeling System (GAMS). The model is solved using a two-step solution strategy. The proposed model enables to explore simultaneously all interconnections within the proposed superstructure in order to find the configuration with the optimal trade-offs between capital and energy costs as demonstrated in this paper for different cases of a milk concentration process.