论文部分内容阅读
由于现代战争的快节奏和异常激烈,在面向服务的军事综合电子信息系统中候选服务的服务质量往往随时间快速变化,有时还有服务的加入和退出,现有组合服务选择方法很难应对这种场景.提出了一种基于危险理论的动态约束多目标免疫克隆算法(DCMOICADT)用于QoS动态变化的服务选择.首先将基于QoS的军事信息服务选择问题建模为带QoS约束的动态多目标组合优化问题,接着采用基于危险理论的动态约束多目标免疫克隆算法同时优化多个目标函数,最终产生一组满足约束条件的Pareto最优解服务组合集.对比实验结果表明,DCMOICADT设计了环境感知因式用于描述QoS动态变化,使用Pareto-占优集和有益不可行解协同的免疫进化方案,能根据当前环境的变化快速且自适应地调整各免疫操作,所得最优解集具有较好的多样性和较强的逼近性,能有效解决QoS动态变化的军事信息服务选择问题.
Due to the fast-paced and furious nature of modern warfare, the service quality of candidate services in service-oriented military integrated electronic information systems tends to change rapidly over time, and sometimes there are additions and withdrawals of services, making it difficult for existing portfolio service selection methods to deal with this This paper proposes a dynamic constrained multi-objective immune cloning algorithm (DCMOICADT) based on danger theory for service selection of dynamic change of QoS.First, the QoS-based selection of military information services is modeled as a dynamic multi-objective Combinatorial optimization problem, and then use the dynamic constraint multi-objective immune cloning algorithm based on danger theory to optimize multiple objective functions simultaneously, and ultimately produce a set of Pareto optimal solution service combinations that meet the constraints.Comparative experimental results show that DCMOICADT designed the environmental perception Factorization is used to describe the dynamic changes of QoS. Using the immune evolutionary scheme coordinated by Pareto-dominant sets and beneficial unfeasible solutions, the immune operations can be quickly and adaptively adjusted according to the changes of the current environment, and the optimal solution set obtained is better Diversity and strong approximation, can effectively solve the dynamic change of QoS military information service selection Choose the question.