CAJ | 학술논문

针对机械加工行业中多任务加工的流水线自动优化控制问题，建立了以“多工件多工序最短加工时间”为目标，以“实际加工限制”为约束的流水线加工控制数学模型，提出一种基于多层编码的遗传-粒子群融合算法。该算法以粒子群算法思想为基础，融合了遗传算法操作和多层编码机制，可防止陷入局部最优解，并具有较快的收敛速度。针对不同规模的加工实例进行仿真，结果表明该算法完成多工件多工序的加工时间比基本遗传算法缩短了约9%，具有提高流水线利用率和机械加工生产效率的优点。
침대궤계가공행업중다임무가공적류수선자동우화공제문제，건립료이“다공건다공서최단가공시간”위목표，이“실제가공한제”위약속적류수선가공공제수학모형，제출일충기우다층편마적유전-입자군융합산법。해산법이입자군산법사상위기출，융합료유전산법조작화다층편마궤제，가방지함입국부최우해，병구유교쾌적수렴속도。침대불동규모적가공실례진행방진，결과표명해산법완성다공건다공서적가공시간비기본유전산법축단료약9%，구유제고류수선이용솔화궤계가공생산효솔적우점。
A genetic-particle swarm optimization algorithm of multilayer-coded is proposed to solve the flow shop’s optimizing control problem of machinery industry. The objective of the issue is to minimize the production time. Through analyzing the actual machining conditions, control mathematical model of production line is set up. The algorithm bases on particle swarm algorithm and combines the genetic swarm optimization with the multilayer-coded mechanism. The algorithm has a higher speed and is used to avoid being trapped into local minima. The results show that the multilayer-coded genetic-particle swarm optimization algorithm has saved 9%times from the basic genetic algorithm. It has advantages of improving utilization ratio and production efficiency in machining line.