Web Service Composition Research Based On Differential Evolution Algorithm

With the maturity and development of cloud computing technology,Web service composition under cloud computing has become a hot spot that attracts more and more attention.The traditional resource pool in the performance of the single Web service is difficult to meet the growing needs of people,the purpose of the Web composition is the reuse of services,Web services,mutual communication and collaboration between the effective integration of various functions of Web services,which To complete and solve some complex and diverse issues and to realize the added value based on the existing services.At present,the most widely used indicator in evaluating Web services is Quality of Service(QoS),which can only reflect the quality characteristics of services and can not reflect users' satisfaction with the quality of service experience.QoE(Quality of Experience)Can well reflect the quality of service users experience.This thesis mainly studies the use of differential evolution algorithm to solve Web service composition problem based on QoE evaluation index.First of all,a Web service composition model based on QoE is studied on fuzzy expert system,and it is transformed into a mathematical model about QoE.Secondly,based on the differential evolution algorithm(DE),a parametric adaptive DE algorithm is proposed,which uses chaos initialization,niche elimination strategy and parameter adaptation mechanism to keep the population diversity in the process of iteration and at the same time to dynamically adjust the crossover rate CR and the scale factor F.Finally,the performance simulation of adaptive DE algorithm and PSO algorithm,ACO algorithm,DE algorithm and chaos initialization niche DE algorithm are carried out.The experimental results show that the parameter adaptive DE algorithm is more effective in solving the problem of Web service composition optimization,Convergence faster,and more stable,but the implementation of time is longer.