计算机科学 ›› 2012, Vol. 39 ›› Issue (Z11): 63-66.

• 服务化的科研成果 • 上一篇    下一篇

基于赋时影响网的模拟退火与粒子群混合改进算法

赵鑫业 唐帅 杨妹 黄柯棣   

  1. (国防科技大学机电工程与自动化学院 长沙 410073)
  • 出版日期:2018-11-16 发布日期:2018-11-16

Hybrid Algorithm Based on Particle Swarm Optimization and Simulated in Timed Influence Nets

  • Online:2018-11-16 Published:2018-11-16

摘要: 赋时影响网是一种最近兴起的分析复杂系统关键事件行为及因果关系的建模范式,通过连接行动事件和期 望效果来实现关系推理过程。基于赋时影响网设计算法在不确定条件下寻找有效行动方案和备选行动方案用以辅助 系统分析决策。本文在标准粒子群优化算法的基础上给出了一种模拟退火改进策略:在规定的计算周期内以粒子群 算法进行进化计算;同时为了避免陷入局部最优,采用模拟退火方法对所有粒子重新进行有选择的初始化,初始化之 后再次应用粒子群算法。为了有效提高算法的运算速度,算法的实现应用了MPI。仿真结果表明,该算法具有较快的 寻优能力和较好的鲁棒性。

关键词: 赋时影响网,行动方案,基于效果作战,模拟退火算法,粒子群算法,MPI

Abstract: The timed influence nets arc an emerging formalism which helps a system modcler in connecting a set of ac- tionable events and a set of desired effects through chains of cause and effect relationships in a complex uncertain situa- lion. The objective is to assess the behavior of the variables of interest as a function of both the timing of actions and the receipt of evidence on indicators, thus providing aid to decision makers in the revision of the planned courses of actions. This paper proposes an Simulated Annealing Algorithm (SA) based approach integrating standard Particle Swarm Opti- mization (PSO) for finding effective courses of action (CO八、).PSO is used in specified period at first, then employs S八 to initialize again for avoiding dropping the local best, and the arithmetic is executed iteratively until the ending condi- lion is satisfied. For the sake of efficiency, we arc going implement the arithmetic using MPI. hhe results suggest the approach appears to work well in uncertain situation.

Key words: Timed influence nets, Course of action, Effects based operations, Simulated annealing, Particle swarm optimization, MPI

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!