LAUSR

Abstract Graphical tools are useful in the heat exchanger network (HEN) retrofit to maximise energy savings. The disadvantage of heuristic rules, which are usually applied to make retrofit decisions using graphical tools, is that they could lead to sub-optimal solutions. The presented study developed a two-stage method for HEN retrofit. In the first stage, a mixed-integer linear programming (MILP) model is formulated based on the structure of the shifted retrofit thermodynamic grid diagram (SRTGD) to minimise the utility cost and investment. The non-linear equations for the investment cost calculation were linearised, and the parameters in the linearised equations were obtained using data regression. In the second stage, a particle swarm optimisation (PSO) algorithm was selected and applied to adjust the inlet and outlet temperatures of heat exchangers with the aim of minimising the payback period on the basis of the first-stage solution. The proposed two-stage procedure combines the strengths of the MILP and PSO methods, offering convenient interfaces for user interaction and results interpretation. Two cases were studied to verify the effectiveness of the method. Case 1 and Case 2 decreased the payback period by 11.6% and 21.7% compared to the results obtained in previous retrofit applications.