带时间窗农机调度问题的改进遗传算法

Improved Genetic Algorithm for Scheduling of Agricultural Mechinery with Time Window

在农机调度问题中,农田作业时间窗是需要解决的重要问题。首先构建了多因素约束的农机调度模型,然后设计了改进的自适应遗传算子的农机调度方法。算法以农田为单位进行染色体编码生成方案,并提出了父代染色体以随机生成的掩码为依据在指定位置交叉的算法规则和由种群状态来确定个体交叉概率与变异概率的自适应方法。最后应用实际案例数据模拟实验比较了本文改进的算法和参考文献中算法的优越性,实验结果表明:当农田维数变化,农机数量确定时,本文算法比文献算法的平均费用降低了6.0%,证明了本文算法的有效性。

Time window of farmland operation is an important problem which needs to be solved in scheduling of agrocultural mechinery.To solve this problem,the multi-factor constrained agricultural mechinery scheduling model was established first,the improved self-adaption genetic arithmetic operator agricultural mechinery scheduling method was designed then.The chromosomo coding was carried out in farmland through this algorithm,based on it,a algorithm rules for parent chromosome cross at a specified location based on a randomly generated mask and a self-adaption nethod for determing individual crossover probability and mutation probability by population state were put forward.The adventages of this improved algorithm were compared with those in referencesthrough actualcasesdata simulation experiment.The result ofthis experiment indicates that,the average cost which adopts this algorithm is 6.0%lower than the average that adopts algorithm in references when the dimension of farmland changes while the number of agricultural mechinery is determined,the effectiveness of the algorithm in this paper was proved.