混沌蚁群算法的Web服务组合优化研究

摘要

In order to satisfy the users' increasing demands on Quality of Experience (QoE) of services,Web service composition based on QoE is proposed.On the basis of Fuzzy Expert System,the mathematical model of QoE applied to Web service composition optimizing problem is put forward.Chaos Ant Colony Optimization (CACO) is used to solve Web service composition.According to the ergodicity,randomness and regularity of chaos,the algorithm adds to the chaos disturbance to avoid falling into local optimal solution and the global optimal solution will be found.Compared with the original Artificial Bee Colony (ABC),Particle Swarm Optimazation (PSO) and Ant Colony Optimization (ACO),the experimental results show that CACO has shorter operating time,faster convergence and high stability in Web service composition problem and has a better developmental prospect.%为保证Web服务组合满足用户对Web服务质量日益增长的需求,提出了基于体验质量(Quality of Experience,QoE)的Web服务组合优化方法,即建立模糊专家系统(Fuzzy Expert System) QoE评估模型,并转化为Web服务组合优化的数学模型,采用混沌蚁群算法(Chaos Ant Colony Optimization,CACO)进行Web服务组合优化求解.该方法利用混沌算法的遍历性、随机性和规律性,通过引入混沌扰动来避免优化过程中出现局部最优解,以期获得服务组合的全局最优解.为验证CACO算法的可行性和有效性,对其与人工蜂群算法(ABC)、粒子群算法(PSo)和原始蚁群算法(ACO)等进行了同步对比实验.实验结果表明,CACO算法相比其他算法具有运行时间短、收敛速度快且稳定性高的优点,具有较好的发展前景.