Simultaneous Heat Exchanger Network Synthesis Involving Nonisothermal Mixing Streams with Temperature-Dependent Heat Capacity

Heat exchanger network synthesis (HENS) is faced with the dilemma of substantial nonisothermal mixing and nonconstant thermal properties of streams against the higher computational efficiency required to solve mathematical programming models with a large amount of variables and high nonlinearity. In this work, by replacing the conventional binary variable in a HENS model, a nonlinear approximation term is developed to indicate the existence of a heat exchanger so as to reduce the amount of variables and constraint as well as enhance the model's computational efficiency. With this approximation, a new HENS model was formulated considering simultaneously nonisothermal mixing and temperature-dependent heat capacity flow rate. All the variables of the model are stated with clear upper and lower bounds. A 10-stream (5 hot process streams and 5 cold process streams) benchmark HENS problem is solved by the developed model. The results show that the proposed model can generate a better cost-optimal heat exchanger network (HEN) with more accurate configuration in relatively short solution time compared with that in the literature.