基于遗传算法的设备布置优化算法

对设备布置问题,建立了多目标优化数学模型.为弥补当前的现场布置遗传算法在变异阶段的不足,将最优个体变异与随机变异相结合,设计了组合变异策略:首先变异最优个体,如果变异出更优的个体,则用新个体替换当前种群的最差个体;如果最优个体变异不成功,则随机选择一个个体执行随机变异.据此,提出了一种改进的遗传算法用于求解设备布置问题.仿真实验证明了组合变异策略能够在明显较短的时间内,取得与随机变异相当的最优布置结果.对比分析进一步验证了该算法的有效性.

Optimization of machine layout based on genetic algorithm

To solve the machine layout problem, a multi-objective optimization model was constructed. And a combination mutation strategy, combined with the best individual mutation and the random mutation, was designed to remedy the defects of the present genetic algorithms for site layout problems. At the beginning of combination mutation, the best individual mutation was executed. If a better individual was generated, the worst individual in current population was replaced by the new one. Otherwise, the random mutation was executed on a random selected individual. Based on the combination mutation strategy, an improved genetic algorithm was also proposed to solve the problem of machine layout. Simulation experiments prove that the combination mutation strategy achieves solutions not inferior to the solutions of the random mutation in obviously shorter time. A comparative analysis further verifies the efficiency of the proposed algorithm.