摘要
基于有效基因路径表示的遗传算法:先建立单颗成像侦察卫星调度问题的约束规划模型。采用路径表示方法,对成像任务调度序列对应的染色体进行编码。选择2个染色体作为2父个体。接着采用顺序交叉算子对2父个体进行交叉操作,得到2子个体。然后采用交换变异算子对2子个体进行变异操作,得到2新子个体,并与两个父个体进行比较,比较依据为适应度函数。根据染色体上有效基因的数量以及对应的成像任务的优先级,计算染色体的适应度函数,选择适应度函数值最大的2个体作为下一次遗传操作的父个体。若迭代次数达最大值,算法结束,否则重复算法步骤。
The genetic algorithm based on effective genetic path establishes the single imaging reconnaissance satellite scheduling model at first, and adopt path presentation method to code the chromosome for imaging task scheduling sequence. Select 2 chromosomes as 2 father generations. Second, use sequence crossing operators to carry put crossing operations for 2 father generations and generates 2 filial generations. Third, carry out variation operation for 2 filial generations based on exchanging variation operators and generate 2 new filial generations. Compare them with 2 father generations based on sufficiency function. According to the quantity of effective gene and priority level of corresponding image task, calculate sufficiency function and select 2 biggest sufficiency functions as the father generations for the next time genetic operation. If the iterative times reaches the max, the algorithm is finished, otherwise repeat the steps.
出处
《兵工自动化》
2008年第10期83-85,91,共4页
Ordnance Industry Automation
关键词
遗传算法
调度
约束规划
成像侦察卫星
Genetic algorithm
Scheduling
Constraint programming
Imaging reconnaissance satellite
