Research On Optimization Algorithm Of Multi-objective Cloud Manufacturing Service Composition Considering Quality And Energy Consumption

As a successful smart manufacturing business model and the main driving force of Industry 4.0,cloud manufacturing has attracted extensive attention from scholars at home and abroad in recent years.However,as cloud-manufactured service providers continue to provide cloud services with similar functions but different qualities,how to choose and combine a large number of cloud services to meet users’ service quality expectations,that is,the cloud service combination problem of service quality perception,has become a problem,become a critical issue in computing service provisioning.This paper firstly studies the quality-aware multi-objective cloud manufacturing service composition optimization problem.Secondly,in order to save energy consumption and protect the environment,it further studies the energy consumption-based cloud manufacturing service composition problem.The cloud manufacturing service composition problem often has multiple optimization objectives,and the weight of each objective is difficult to determine in advance.In order to significantly improve the population diversity of the evolutionary algorithm in the solution process and effectively balance the global and local search capabilities of the evolutionary algorithm,this paper proposes a method based on Evolutionary algorithms for adaptive selection and reverse learning strategies,while optimizing for time,cost,reliability,availability,and credibility.First,in order to shorten the solving time of the problem,the K-means method is used to cluster the candidate services based on the service quality,and the services with poor quality are removed in each cluster of candidate services,and these invalid candidate services are skipped;secondly,after the population initialization The reverse learning strategy is used in the stage to make the initial population more widely distributed in the target space and improve the diversity of the population;then,in the algorithm iteration process,every certain algebra,dynamically adjust the selection probability according to the contribution of the new solution to effectively balance the algorithm Finally,the strategy verification and algorithm comparison experiment verification is carried out,and the results show that the proposed algorithm has better comprehensive performance.Under the pressure of increasing energy costs and increasingly prominent environmental concerns,manufacturers need to seek economical solutions that are energy-efficient and low-carbon.Aiming at the above problems,this paper studies the optimization problem of service composition based on energy consumption.First,the corresponding problem formulations are proposed,one objective is to maximize the overall Qo S performance and minimize Qo S risks to obtain higher Qo S,and the other objective is to minimize logistics energy consumption and manufacturing energy consumption to achieve higher Qo S low energy consumption;secondly,a PC-NSGA-II algorithm based on preference crowding distance is proposed,which integrates the user’s preference into the calculation of crowding distance,so that the population searches for the area of interest to the user to a certain extent.The crossover operation of the meta-crossover operator improves the convergence performance of the algorithm,and the mutation probability that changes with the increase of the number of iterations is adopted,so that the population still maintains a certain search ability in the later stage of evolution.Finally,experiments verify the effectiveness of our proposed algorithm.