基于聚类与时间耦合执行序列的任务分解方法

Task Decomposition Method Based on Clustering and Time Coupling Execution Sequence

针对复杂任务分析中的高度耦合任务不易分解且需重构排序的问题,提出了一种基于聚类分析与改进时间-耦合执行序列的自适应任务分解方法;在矩阵最值遴选模型和任务序列转移策略相结合的基础上,设计了基于中间任务序列的任务矩阵分割算法,基于中间任务序列的任务矩阵分割算法很大程度上能够灵活处理离散任务序列并入的问题;并进一步采用粒度自主循环调整机制,此机制能够有效的解决任务粒度分解不均问题,并不断的进行粒度调整重置,从而使得任务粒度自主分解,最终实现了复杂任务的自适应解耦分析;仿真验证结果表明,该方法能够有效的在实现任务粒度自主设计的基础上,进一步实现复杂任务的解耦及序列重构,在作战任务分析领域具有很好的推广应用前景。

Aimed at the problem that highly coupled tasks in complex task analysis are not easily decomposed and need to be reordered,an adaptive task decomposition method based on the cluster analysis and improved time-coupled execution sequences is proposed.Based on the combination of the matrix optimal selection model and task sequence transfer strategy,a task matrix partitioning algorithm based on intermediate task sequences is designed to largely handle the problem of merging discrete task sequences flexibly;The autonomous circular adjustment mechanism of granularity is further adopted to effectively solve the problem of uneven task granularity decomposition and continuously reset the granularity adjustment,thus making the task granularity decomposition autonomously,and finally realizing the adaptive decoupling analysis of complex tasks.The simulation results show that on the basis of the autonomous task granularity design,the method can effectively achieve the decoupling and sequence reconstruction of complex tasks,and it has a very good prospect of application in the field of combat mission analysis.