基于入侵杂草蝙蝠双子群优化的装备保障编组协同任务规划

Cooperative task scheduling for equipment support groups using invasive weed bat dual-subpopulation optimization algorithm

针对装备保障编组协同任务规划问题,构建以时效优先为目标,考虑保障任务时序逻辑关系、任务执行质量和保障编组占用冲突等复杂约束以及保障编组能力更新机制的数学模型,提出一种基于入侵杂草蝙蝠混合算法的双子群任务规划方法。首先,采用佳点集初始化方法,在解空间生成具有均匀分布特征的种群;其次,设计具有修复操作的解编码和任务优先排序,实现任务-编组-时间的匹配和冲突消解;再次,划分双子群,利用入侵杂草优化算法和Fuch混沌蝙蝠优化算法协同进化;最后,应用重组算子引导种群进化,均衡算法全局探索和局部搜索能力。仿真算例表明,所提方法可对大规模复杂任务分配方案进行精确高效的求解。

This paper proposes a hybrid task scheduling method based on the invasive weed bat algorithm upon dualsubpopulation to achieve objective of minimal task implementation time for equipment support groups cooperative task scheduling. The mathematical model is constructed considering complex constrains such as task sequential logical relationship, implementation quality, occupancy conflicts and capability renewal mechanism of support groups. Firstly,the good-point set theory is used to generate initial population with uniform distribution in the solution space. Then,the strategies of repair operator encoding and task priority ordering are designed to achieve tasks-groups-time sequential matching and conflict resolution. Moreover, all individuals are divided into dual-subpopulation by cooperative evolution using the invasive weed optimization algorithm and the Fuch chaotic bat optimization algorithm. Finally, the recombination operator for guiding population evolution is given to balance the exploration and exploitation ability. The simulation example is given to illustrate that the proposed method has better robustness and solving precision for the complicated task allocation scheme.