Abstract
The natural gas liquefaction process occupies a considerable part of the energy consumption in the LNG industry chain, and efficient optimisation methods should be developed to improve the economic and environmental performances of this process. In this work, a new optimisation strategy based on the particle swarm optimisation (PSO) algorithm and bisection method was developed to optimise the propane precooled refrigerant process (C3MR). In the proposed optimisation strategy, the molar fractions of the mixed refrigerant (MR) components were directly adopted as variables in the PSO algorithm, and the MR flow rate was independently adjusted by the bisection method to satisfy the minimum temperature approach constraints. The results show that the optimal operating conditions reduced the energy requirement of the C3MR process by 4% and decreased the consumption of MR by 31.9%, compared to traditional using the flow rates of MR components as variables. The correlations among the properties of key streams were calculated based on the intermediate results of the proposed method and visualised with the heat map. The results show that the energy requirement of the C3MR process decreased with the increase of the low boiling temperature components and increased with the high boiling temperature components of MR.
•A new strategy based on PSO algorithm and bisection method was proposed.•Directly taken the molar fractions of MR components as decision variables.•The energy consumption of the C3MR process was reduced by 4%.•The requirement of MR was decreased by 31.9%.•Using a heat map to analyze the correlations among the properties of key streams.