Cloud Manufacturing Service Composition Research Based On Improved Genetic Algorithm

As a new service-oriented manufacturing model,cloud manufacturing provides users with various manufacturing resources through virtualization and service of manufacturing resources and manufacturing capabilities.These resources can be accessed at any time and paid for as a cloud service.In order to meet the complex manufacturing needs of users,simple cloud services can be grouped into complex cloud services through service composition.Complex cloud services execute in a distributed,heterogeneous,and autonomous environment to accomplish highly uncertain and dynamic manufacturing tasks.Selecting the best combination of cloud manufacturing services to perform manufacturing tasks is one of the key issues to realize resource value-added in cloud manufacturing environment.In this paper,cloud manufacturing service composition optimization is taken as the research object.Based on the service quality evaluation model,the attribute value of cloud manufacturing service composition optimization is fuzzy processed,and the in-depth study is carried out,mainly including the following contents:(1)The Qo S evaluation model was established,and the Qo S attributes were fuzzy and normalized on the basis of cloud model theory,and four evaluation indexes were obtained,including time,cost,reliability and satisfaction.According to the characteristics of service composition optimization problem,a single objective optimization method is selected.This method is to construct the objective function according to the weighted combination of the user's preference features,transform the multi-objective optimization problem into a single objective optimization problem,and then use the single objective optimization algorithm to find the optimal solution of the objective function.(2)A multi-objective optimization algorithm based on improved differential evolution algorithm and genetic algorithm is proposed.In the initial stage of the population,the diversity of the initial population was improved by introducing individual dissimilarity factors to avoid the crossover operation of similar individuals.The improved roulette method was used to select the individuals with higher fitness values in the population.In the early stage of evolution,the value with higher crossover probability is selected to give full play to the global search ability of genetic algorithm.In the later stage of evolution,the value with lower crossover probability is selected to improve the local search ability.In the mutation stage,differential evolution algorithm is used to carry out mutation operation.Finally,IDGA algorithm is used to simulate the Qo S evaluation parameters.The experimental simulation results show that compared with the basic genetic algorithm and differential evolution algorithm,IDGA alorithm's task execution time and execution cost are significantly better than the genetic algorithm and differential evolution algorithm.(3)The concrete steps of IDGA algorithm used to solve the problem of cloud manufacturing service composition are given.The improved IDGA algorithm is combined with a complex manufacturing task to simulate the optimization problem of cloud manufacturing service composition.Finally,the performance of the algorithm is analyzed from four aspects: convergence,effectiveness,stability and execution time.The experimental data show that,compared with basic genetic algorithm,differential evolution algorithm and ant colony algorithm,the improved IDGA algorithm can solve the multi-objective optimization problem of cloud manufacturing service composition more efficiently.The IDGA algorithm designed in this paper can not only ensure the diversity of the initial population,but also has the advantages of strong global search ability of genetic algorithm.After the introduction of differential evolution algorithm in the mutation stage,it also has the advantages of strong local search ability and fast convergence speed.