基于混合遗传算法的飞机腹舱装载优化  

Optimization of Aircraft Cargo Hold Loading Based on a Hybrid Genetic Algorithm

作　　者：张长勇 王彤 ZHANG Changyong;WANG Tong(College of Electronic Information and Automation,Civil Aviation University of China,Tianjin 300300,China)

机构地区：[1]中国民航大学电子信息与自动化学院,天津300300

出　　处：《包装工程》2024年第21期200-207,共8页Packaging Engineering

基　　金：中央高校高水平培育项目(3122023PY04)。

摘　　要：目的以航空运输背景下货舱行李的码放问题为研究对象,旨在实现货舱空间利用率的最大化。方法在考虑行李姿态、货舱空间、体积、重量、不重叠等多种实际约束的条件下,搭建航空行李码放模型,设计一种启发式与改进遗传算法相结合的混合遗传算法。首先利用基于极值点的启发式算法对初始种群加以优化,加快种群进化速度,得到初始高质量装载方案;然后通过设计混合选择及混合变异算子,利用改进的遗传算法对多组可行方案寻优,进而得到货舱空间利用率最大、垛型重心偏移量最小的装载布局方案;最后通过某机场自助行李托运系统采集的真实旅客行李数据进行实验,并实现装载方案的可视化。结果利用偏大、偏小和混合型3种行李数据进行实验测试,实验结果表明在满足货舱多种实际约束的条件下,混合遗传算法平均空间利用率达到89.71%,相比已有货舱码垛算法提高了9.25百分点,垛型规划更加科学合理。结论所提算法可快速收敛并防止局部最优,对于不规则货舱空间以及异构行李具有更好的适应性,可实现货舱码垛的快速合理布局规划。The work aims to study the baggage stacking in the cargo hold under the background of air transportation,so as to maximize the utilization of cargo hold space.Considering practical constraints such as baggage posture,volume,weight,non-overlapping requirements,a model for aviation baggage storage was developed.Additionally,a hybrid genetic algorithm combining heuristic techniques with an improved genetic algorithm was designed.Firstly,an extreme points-based heuristic algorithm was employed to optimize the initial population and accelerate the population evolution speed to obtain high-quality initial loading schemes.Then,through the design of hybrid selection and mutation operators,the improved genetic algorithm was used to optimize the multiple feasible schemes,resulting in a loading layout scheme with maximum cargo space utilization rate while minimizing deviation of palletizing center of gravity.Finally,an experiment was conducted with real passenger baggage data collected from an airport self-service baggage check system to visualize the loading scheme.The baggage data in oversized,undersized,and mixed types were selected for experimental testing.The results demonstrated that under the conditions of satisfying various practical constraints of cargo holds,the average space utilization rate of the hybrid genetic algorithm reached 89.71%.This represents an improvement of 9.25%compared to existing cargo hold stacking algorithms,resulting in more scientific and reasonable stacking planning.The proposed algorithm can converge quickly and prevent local optima,showing better adaptability to irregular cargo hold spaces and heterogeneous luggage.It enables rapid and reasonable layout planning for cargo hold stacking.

关 键 词：三维装箱 航空行李 混合遗传算法 多约束 

分 类 号：TB485.3[一般工业技术—包装工程] TP301.6[自动化与计算机技术—计算机系统结构]