Industrial internet edge service cache decision-making method and system

Abstract

The invention relates to the technical field of industrial internet, in particular to an industrial internet edge service cache decision-making method and system. According to the industrial internet edge service caching decision-making method provided by the embodiment of the invention, through a service caching method based on deep meta reinforcement learning, the perception ability of deep learning, the decision-making ability of reinforcement learning and the rapid environment learning ability of meta learning are combined; the fused framework can quickly and flexibly obtain an optimal caching strategy from a dynamic environment. The corresponding system also has the same technical effect.

Description

technical field [0001] The present invention relates to the technical field of industrial Internet, in particular, to a decision-making method and system for edge service caching of industrial Internet. Background technique [0002] By integrating advanced technologies such as 5G communication and artificial intelligence, the Industrial Internet integrates various sensors and controllers with perception and control capabilities into the industrial production process to optimize product production processes, reduce costs, and increase productivity. Due to the characteristics of centralized deployment, the traditional cloud computing model usually has computing nodes far away from intelligent terminals, which makes it difficult to meet the needs of high real-time performance and low latency in the industrial field. By sinking computing, storage, and network resources to the edge of the industrial network, edge computing can respond to device requests more conveniently, meet ke...

AI Technical Summary

Problems solved by technology

However, because content popularity and user preferences change over time and are unpredictable

At the same time, in the face of fast-changing industrial application scenarios, every time the environment changes, we have to readjust the caching decision of the service through recalculation, otherwise high service delay and cost will be generated

Although some good results have been achieved in edge cache decision-making by introducing intelligent algorithms such as deep learning and reinforcement learning, there are still challenges such as slow learning speed and invalidation of original network parameters when the model environment changes.

Images