基于多子群目标分段差分进化的多目标热连轧负荷分配

Multi-objective load distribution of hot strip mills based on multi-swarm and sub-objective differential evolution

摘要: 针对热连轧负荷分配的目标函数处理中采用的加权系数不好确定的问题,设计了一种多子群目标分段优化的差分进化算法.每个子群优化目标函数的一个子目标,子群之间独立进化,不但解决了加权系数问题,并且提高了算法的收敛速度和精度.最后,算法的有效性通过仿真进行了验证.

Abstract: A differential evolution algorithm based on multi-swarm and sub-objective optimization is presented in order to solve the difficulty in selecting the weighting coefficients in processing the objective function of hot strip mills.Each sub-swarm optimizes a sub-objective and evolves independently.This not only solves the issue of weighting coefficients,but also increases the convergence speed and accuracy.Finally,the algorithm's effectiveness is verified by simulation.