Research On Combination Optimization Of Cloud Manufacturing Service Based On Ant Colony Algorithm

The manufacturing industry is an important pillar of the national economy and national security,and is a country with a strategic position.The manufacturing industry is developing towards diversification,complexity and large-scale.Now,a more efficient and secure manufacturing mode is urgently needed to adapt to the change of demand.The concept of Cloud Manufacturing came into being.Cloud manufacturing is an Internet based,service-oriented,efficient and low-power new manufacturing mode.Its implementation depends on the development of technology such as manufacturing digitalization,cloud computing,Internet of things,big data and so on.In the cloud manufacturing platform system,the distribution of manufacturing resources will change dynamically with the different needs of the user.When a node fails,the platform can transfer tasks to other manufacturing resources when users fail to know it,so the cloud manufacturing mode has higher safety and reliability compared with the traditional manufacturing mode.In order to better solve the problem of cloud manufacturing service composition optimization,this thesis proposes an ant colony algorithm(Dynamic Parameter Ant Colony Algorithm,DPACO)based on Quality of Service(QoS)evaluation model.First,a multi attribute QoS evaluation model based on Time,Cost,performance and satisfaction is established.Secondly,the dynamic parameters of ant colony algorithm,the algorithm is divided into two stages,the first stage of the selection heuristic factor is small,the larger the expected heuristic factor beta and pheromone evaporation coefficient,and adding special ant;second stage selection heuristic factor is larger,the smaller the expected heuristic factor beta and pheromone evaporation coefficient,and join the subspace search mechanism.Finally,the task of steel forging simulation experiment,the DPACO and the basic ant colony algorithm(ACO),genetic algorithm(GA),differential evolution algorithm(DE)for comparison,the experimental results show that DPACO has better in the solution of cloud manufacturing service composition problem of convergence,effectiveness and robustness,but in the algorithm execution time is slightly more than the other basic algorithm.