改进平衡优化器算法求解柔性车间调度问题

An Improved Equilibrium Optimizer Algorithm for Flexible Job Shop Scheduling Problem

针对原始平衡优化器算法(Equilibrium Optimizer,EO)求解车间调度问题时出现精度低、稳定性差的问题,提出一种基于单向多种群信息交流的量子改进平衡优化器算法(Improved Multipopulation Quantum Equilibrium Optimizer,IMQEO)。首先,将初始化平衡池分为三个子平衡池,一个平衡池主要肩负开发功能,其余平衡池主要用于空间探索,以高效搜寻最优解;然后分离最优浓度各分量,重建多个最优浓度,结合贪婪策略,个体依次包围收缩于各最优浓度,实现加速收敛;最后,利用量子旋转门策略更新浓度以跳出局部最优解。对比标准EO算法,经过车间调度算例测试,结果表明混合改进策略具有较好的优化效果。

Aimimg at low accuracy and poor stability of the original equilibrium optimizer algorithm in solving the job shop scheduling problem,a improved multipopulation quantum equilibrium optimizer(IMQEO)based on unidirectional multi-popula-tion information exchange is proposed.Firstly,the initial balance pool is divided into three sub-balance pools.One balance pool is mainly used for exploitation,and the other balance pools are mainly used for space exploration to find the optimal solution efficient-ly.Then,the components of the optimal concentration are separated,and multiple optimal concentrations are reconstructed.Com-bined with the greedy strategy,the individuals are successively surrounded and contracted to each optimal concentration to realize accelerated convergence.Finally,a quantum revolving door strategy is used to update the concentration to jump out of the local opti-mal solution.Compared with EO,the results show that the hybrid improvement strategy has a better optimization effect.