An improved multi-objective particle swarm optimization algorithm based on quantum behavior for an ethylene cracker

The multi-objective particle swarm optimization based on quantum behavior algorithm (MOQPSO) has a great improvement in performance compared with the conventional PSO-based approach for optimizing multiple objectives, but the MOQPSO still has the disadvantages of low global convergence ability and prone to getting trapped in local optima during optimization in the late period. Therefore, an enhanced particle swarm method for optimizing multiple objectives based on quantum behavior algorithm (IMOQPSO) is proposed. Based on the improved delta well model, the probability of particles appearing far away from the central point is increased, the particle search scope is expanded, and the proposed algorithm exhibits enhanced global convergence capabilities. At the same time, the triple dynamic switching of characteristic length improves the adaptability of the proposed algorithm. Finally, the IMOQPSO algorithm is compared with other prevalent multi-objective optimization methods on the test function, and the results are compared to verity that the IMOQPSO has better capability to converge globally and aptitude for local exploration. Furthermore, both the IMOQPSO and the reference algorithm are employed to optimize the ethylene cracking furnace model, using ethylene yield and propylene yield as the target objectives. The results show that the IMOQPSO achieves the best results compared with the initial operational parameters during the optimization process of the main product yield under the fixed cycle regulation operating conditions.