Artificial Hummingbird-Based Optimisation with Advanced Crowding Distance of Energy Reduction in the Polyethylene Reactors

Ethylene is polymerised by free radicals under extreme conditions of high pressure and temperature to produce low-density polyethylene LDPE. Considering the requirement for high compression power and heating–cooling elements, combined with depleting fossil fuel and climate change issues, an approach is needed to trade-off these issues. As such, an effective approach of multi-objective optimisation study to obtain the optimum production of the LDPE with minimum energy consumption is proposed in this work. The multi-objective artificial hummingbird algorithm with dynamic elimination-based crowding distance (MOAHA-DECD) executes within ASPEN Plus–MATLAB environment for energy saving of low-density polyethylene (LDPE) production. Three problems are addressed: minimise energy cost and maximise productivity for problem 1 (P1); minimising energy cost and maximising conversion for problem 2 (P2); and minimising energy cost, maximising productivity, and maximising conversion for problem 3 (P3). The inlet pressure, the mass flow rate of Initiator 1 (tert-butyl peroxypivalate, TBPPI), and the mass flow rate of Initiator 2 (tert-butyl 3,5,5trimethyl-peroxyhexaonate (TBPIN)) of the reacting zones (zone 3 and zone 5) are considered as decision variables. Pareto solutions obtained are arrayed across the entire Pareto front (PF) with an even sweep and diverse points. Based on the results, the highest productivity, lowest energy cost, and highest conversion are 554.958 Mil. RM/year, 61.388 Mil. RM/year, and 0.320. The decision variable plots show that the mass flow rate of the initiator at the end zone of the reactor highly impacts the optimal option. For the next study, the generated Pareto allows decision-makers to select the most acceptable solution based on their preferences to trade-off economic, energy, and environmental issues.