An Improved MOPSO Algorithm for Operation Optimization of Ethylene Cracking Furnace

The traditional MOPSO algorithm has the disadvantage that the convergence is poor when the multi-objective optimization task is processed, and the locally optimal solution is easy to fall into the local optimal solution. Therefore, an improved MOPSO algorithm is proposed. A double search strategy is adopt to improve the convergence, where the particles learns the individual optimal position in the early iteration and learns the global optimal position in the late iteration. A migration mechanism is used to improve the communication ability of the particles in the external archive. Finally, the proposed MOPSO algorithm is used in the operation optimization of ethylene cracking furnace. The experimental results show that the improved MOPSO algorithm can obtain higher ethylene yield than the traditional MOPSO algorithm using the feedstock flow, the outlet temperature, the ethylene yield and the propylene yield as the optimization targets.