基于混合遗传算法的多头动臂式贴片机贴装过程优化研究

摘要

The component placement process of multi-head gantry mount machine was analyzed. The optimization problem of placement process was decomposed to three sub-problem of component grouping,component group picking and placing. The component arrangement in slots was a critical factor in solving the optimization problem of placement process of the multi-head gantry mount ma-chine. Based on the analysis of practical engineering application,with the minimum move distance of arm as an objective,an optimiza-tion model of the component placement process of multi-head gantry mount machine was formulated. Combined with heuristic method and Genetic Algorithm (GA)method,the hybrid GA (HGA)was adopted to optimize the placement process. A two-dimensional real number coding method was proposed in the HGA. In the individual chromosome,the feeder number and slot number were used to de-scribe the arrangement sequence and position of component types. By aiming at the proposed coding method,the improved order cross-over and adaptive mutation were used,and a parallel structure was contained in the HGA. By integrating local search strategy,the component placement sequence and the feeder arrangement are optimized simultaneously. Through calculation examples and results comparison,it shows that this method can achieve optimization of the component placement process of multi-head gantry mount machine and improve the assembly efficiency.%对多头动臂式贴片机的贴装过程进行了分析，将其贴装过程优化问题分解为元件分组、元件组拾取和元件组贴装3个子问题。元件在供料槽中的布置是求解多头动臂式贴片机贴装过程优化问题的关键。在分析实际工程应用的基础上，以动臂的移动距离最小为目标，建立了多头动臂式贴片机的贴装过程优化模型。用启发式方法和遗传算法（GA）相结合的混合遗传算法（HGA）实现了贴装过程的优化。在HGA中提出了一种二维实数编码方法，用供料器编号和供料槽编号描述元件类型在供料槽中的位置和顺序。针对提出的编码方式，采用了改进的顺序交叉和自适应的变异操作，并在算法内采用了并行结构，结合局部搜索策略，实现了元件贴装顺序和供料槽布置同时优化。通过实例计算和结果比较表明该算法能实现多头动臂式贴片机贴装过程的优化，提高印刷电路板（PCB）装配效率。