基于记忆曲线的ACO在柔性作业车间的调度优化

Scheduling Optimization of Ant Colony Algorithm in Flexible Job Shop Based on Memory Curve Model

为克服蚁群算法存在收敛速度慢、容易陷入局部最优解的问题,通过研究记忆曲线模型和蚁群算法信息素更新规则的特点,提出了一种基于生物记忆曲线模型的信息素更新规则对蚁群算法进行改进,并通过实验确定改进后的蚁群算法各参数的合理取值。以最短加工时间为目标函数,建立柔性作业车间调度的目标函数,结合实际算例借助MATLAB求解。通过与其他改进蚁群算法的对比,对6个Job-Shop Benchmark的基准问题进行仿真,通过仿真结果发现,无论是最优解的质量还是求解速度上改进的蚁群算法较基本蚁群算法都有较大提升。最终得出本文提出的基于生物记忆曲线模型的信息素更新规则具有良好的求解能力和收敛能力。

In order to overcome the problem of convergence slow, and easily falling into local optimal solution in basic ant colony algorithm, by studying the characteristics of memory curve model and the pheromone updating rule in ant colony algorithm, And algorithm reasonable value of the parameters an improved ant colony algorithm has been proposed. is determined by the improved ACO by experiments. Taking the shortest processing time as the objective function, the paper combines with practical examples using MATLAB to solve. Compared with other intelligent algorithms, the paper simulates benchmark issues of the 6 Job - Shop, and simulation results show that, compared with basic ACO, both the quality of the optimal solution and the speed of solving the problem have greatly improved. Finally, pheromone up- dating rule based on biological memory curve model has better solving ability and convergence capability.