Research On Multi-gent System Scheduling In Cloud Manufacturing Environment

With the rapid development of emerging technologies such as cloud computing,big data,Internet of Things,artificial intelligence and the rapid integration of manufacturing and information technology,the use of emerging technologies such as cloud manufacturing and multi-agent systems has been applied to the manufacturing industry.The application of cloud manufacturing technology strengthens the production cooperation between enterprises,achieves decentralized manufacturing resources centralized use,centralized manufacturing resources scattered services.However,in the existing cloud manufacturing platform,the task and resource scheduling process does not take the willingness of resource owners and the availability of intermittent resources into account.The real-time changes in idle resources and the production of enterprises in the cloud manufacturing platform are facing diversification and individualization.In order to solve the above problems,this paper combines the idea of cloud manufacturing with the idea of multi-agent system,puts forward the multi-agent system in cloud manufacturing environment,and research it in detail.The main research work is as follows.a.Construct the multi-agent system in cloud manufacturing environment.For the first time,it combined idea of cloud manufacturing with idea of multi-agent system,the concept of cloud agent and multi-agent system in cloud manufacturing environment is put forward,and the overall structure of multi-agent system in cloud manufacturing environment is expounded.The system is divided into multi-agent system based on enterprise private cloud and multi-agent system based on public cloud.The basic structure,operation mode and scheduling idea of the two systems are analyzed respectively.On this basis,the multi-agent system and the scheduling system in the cloud manufacturing environment are constructed.b.Aiming at the internal scheduling problem of the enterprise,taking the parallel machine scheduling problem of two competing task agents as an example,this paper analyzes the scheduling mechanism for this problem of multi-agent system based on enterprise private cloud,and proposes a single distribution strategy for dynamic change of tasks and a centralized distribution strategy for task stability,to achieve task and machine allocation scheduling.c.Aiming at the scheduling problem of inter-enterprise collaboration,it puts forward scheduling mechanism of multi-agent system based on public cloud;consider the intermittent availability of resources,setup time and cost and the enterprise transport time and cost.With the minimum completion time and minimum total cost as the goal,the task and resource scheduling model is established,and the genetic algorithm is used to solve the model.d.For the parallel machine scheduling model of two competing task agents for multi-agent system based on private cloud,objective of two task agents is to minimize the total delay time and minimize makespan,respectively.This article uses the example to verify the feasibility of the single distribution strategy and the centralized distribution strategy.For the task and resource scheduling problem of multi-agent system based on public cloud,we use the example to verify the correctness of the model.e.Finally,the achievement of this article is summarized,and primary plans for the further work are proposed.