基于遗传退火算法的飞机定检原位工作流程优化  

Optimization of Plane's Primary Periodic Maintenance Workflow Based on Genetic Annealing Algorithm

作　　者：吕晓峰[1] 谢勇 席建峰 张勇亮[4] 

机构地区：[1]海军航空工程学院兵器科学与技术系 [2]解放军91224部队 [3]解放军91181部队 [4]海军航空工程学院研究生管理大队

出　　处：《计算机与现代化》2012年第7期25-29,共5页Computer and Modernization

摘　　要：将遗传算法(GA)应用于飞机定检原位工作流程优化中。首先,建立原位工作流程优化模型;其次,提出"排序调整法"来保证个体对应解符合工序约束;最后采用精英选择算子。模拟退火算子和自适应机制对基本遗传算法(SGA)进行改进。仿真结果表明,改进遗传算法在最优解搜索能力上较SGA有明显提高,克服了其容易"早熟"的不足;优化后原位工作完成时间较优化前缩短19.78%,验证了GA在解决定检工作流程优化问题上的适用性。Genetic algorithm（GA） is used to optimize plane＇s periodic maintenance primary workflow. Firstly, the model of pri- mary work is built. Secondly, the way of ＂adjusting the sequence＂ is proposed to ensure the solution of the individuals up to the limits of the work sequence. Finally, elitist operator, simulated annealing（SA） operator and adaptive mechanism are used to im- prove Simple Genetic Algorithm（SGA）. The simulation results demonstrate that, the improved GA is much stronger in best-solu- tion search ability than SGA, and it overcomes its deficiency of being easy to ＂precocity＂ ; after optimization the finish time of primary work is shorter 19.78% than before, and proves that GA is good for the optimization of primary periodic maintenance workflow.

关 键 词：GA 飞机定检 原位工作 流程优化模型 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]