研究并改进了基于双种群遗传算法的矩形优化排样问题的求解方法。使用不同方式产生初始种群,对不同个体使用不同的具有自适应能力的交叉算子和变异算子,使算法的全局优化能力得到提高。以该算法为基础开发了一个应用于实际生产的智能排样系统,对比文献中的数据进行验证,结果表明在原材料利用率方面本方法高于其他类似的正交排样算法。 This paper studies and improves the solving method of rectangular optimization nesting problem based on double population genetic algorithm. Using different ways to generate the initial population, using different adaptive crossover operators and mutation operators for different individuals, the global optimization ability of the algorithm is improved. Based on this algorithm, an intelligent nesting system is developed and compared with the data in the literature. The results show that this method is superior to other similar orthogonal nesting algorithms in terms of raw material utilization.