Mathematical Programming Approach to Total Site Heat Integration

Abstract Heat Integration is one of frequently used methods for decreasing utility consumption. Originally it was developed for integration at the process level and gradually extended to heat recovery between various processes. In this current work a mixed-integer nonlinear programming (MINLP) model has been enhanced to evaluate the rate of heat recovery between those various processes and also to obtain the optimal temperature for the intermediate utility. This is achieved by considering: i) Heat losses during the transporting of steam through the pipeline, ii) Increased investment into heat exchangers and pipelines due to higher pressures and iii) Optimising the areas of heat the exchangers not only at the process level but also for those heat exchanger areas for heat transfer between process stream and intermediate utilities. A created superstructure considers that processes are connected to each other through various intermediate utilities, the temperature of which vary within their temperature intervals. Two different strategies for obtaining Total Site Heat Exchanger networks of a Total Site have been applied: a sequential and a simultaneous one. According to the first strategy Heat Integration is first obtained at the process level and then at the Total Site level from the resulting utility requirements obtained from the first step. In the second strategy the heat recovery is achieved simultaneously at the process and the Total Site levels. Both strategies were compared in order to evaluate, which one performs better.