Research On Optimization Method Of Edge Computing Resource Allocation Strategy For Industrial Internet Of Things

The Industrial Internet of Things has been rapidly developed under the impetus of the intelligent manufacturing industry,and the demand for resources of industrial equipment has also been further expanded.How to alleviate the demand for industrial resources in the face of a huge industrial network has become an urgent problem in the industrial field.In order to meet the different resource requirements of different industrial applications in a complex industrial environment,and to further achieve load balancing.As a new computing generic,edge computing can make full use of the resources on the industrial edge,and combine the decision-making capabilities of reinforcement learning to provide an efficient resource allocation plan for industrial applications.Therefore,this article is based on reinforcement learning for the industrial Internet of Things.The resource allocation strategy has been studied,and the main tasks include:1.Research on the key technologies of Industrial Internet of Things and edge computing,consider the requirements for resources in different industrial application scenarios,and combine the advantages of the architecture based on Software Defined Industrial Internet of Things,and designed a SDIIoT-based The edge computing network architecture provides a basic platform for the efficient allocation of resources.2.Aiming at the problem of the reinforcement learning algorithm performing invalid actions in the process of implementing resource allocation strategies,a preliminary resource selection plan for computing tasks is constructed,which can initially screen the entire network resource plan,and the scope of the selection includes the network responsible for link forwarding.Resources and computing resources that provide computing power.3.Considering that in the process of resource allocation,the complex and changeable industrial environment will reduce the quality of allocation,a resource allocation algorithm based on deep reinforcement learning is designed,which uses the powerful decisionmaking ability of reinforcement learning in complex scenarios to enable different computing tasks Obtained a more efficient resource allocation plan.4.An experimental simulation platform was built to verify the rationality of the above algorithm.The simulation results show that the resource plan proposed for the primary selection of computing task resources basically meets the differentiated requirements of computing tasks for resources;the resource allocation algorithm based on deep reinforcement learning improves the utilization of resources on the edge side and further realizes load balancing.The research results of this article have important reference value for edge computing resource allocation under IIoT.