【摘要】 提出了一种新的基于个体寻优策略的模块划分方法,从模块聚类的内在驱动因素出发,建立相应的规则约束个体的聚合与分离。与遗传算法相比,该算法无需构建目标函数和多次迭代就能获得更好的划分结果,操作流程简洁,容易收敛至全局最优解,计算时间缩短。对比了其他文献提出的算法,表明该算法在解决模块划分问题时的有效性和通用性。更多还原

【Abstract】 Modular design is an effective way to resolve the contradiction between the development cycle,cost and product type of unmanned ground weapon station. Module partition is a key step in modular design. Numerical algorithms have the disadvantages of long calculation time,complicated operation,and difficulty in converging to the global optimal solution. Therefore,a new module partition method based on individual optimization strategy was proposed. Starting from the inherent driving factors of module clustering,rules were established to guide the aggregation and separation of individuals. Compared with the genetic algorithm,the algorithm does not need a long iteration to obtain a better partition scheme. At the same time,the module partition algorithm proposed in other articles were compared,which shows the effectiveness and generality of the algorithm in solving the module partition problem.更多还原