Research On Cloud Edge Collaborative Computing Based Fault Diagnosis System

With the development of the Internet of Things,Sensor technology and Intelligent Manufacturing,ensuring the stable and continuous operation of manufacturing equipment is a key factor to improve the competitiveness of manufacturing companies.Therefore,timely and accurate diagnosis of equipment failures and troubleshooting are of great economic significance for manufacturing companies.The cloud platform has powerful computing and storage capabilities,which is an excellent tool for fault diagnosis.The edge platform is close to data source,which can process and analyze data locally,improving real-time performance,and reducing unnecessary waste of resources resulting from data transmission.Therefore,the combination of cloud computing and edge computing provides a good prospect for real-time,efficient and low-cost fault diagnosis system.Based on this,this paper studies the cloud edge collaborative computing based fault diagnosis system,focusing on the diagnosis Bayesian network model and service composition optimization method in the cloud edge collaborative computing scenario.Finally,the system is designed and implemented,and the feasibility of the system is verified by the actual diagnosis case.First,based on the cloud edge collaborative computing connotation and fault diagnosis scenarios,this paper proposes a layered framework for fault diagnosis system,which integrates different levels of computing resources in a smart factory.Then,based on the characteristics and advantages of cloud computing and edge computing,the modeling of diagnostic Bayesian network is divided into two parts:model training and fault diagnosis,which are separately completed on the cloud and edge.In addition,in order to improve the accuracy of the fault diagnosis results,this paper adds an additional layer of information on the basis of the Naive Bayesian network to merge human observation information,system maintenance information,abnormal operation records,etc.which are related with the probability of fault occurrence.Since the edge node has limited resources,it is not enough to accommodate multiple tasks to be executed at the same time,so next,this paperstudies the optimization problem of edge-side service composition,and focuses on specific fault diagnosis scenarios,transforming traditional resource scheduling optimization into Qo S-based service composition optimize the problem,and use particle swarm algorithm to solve the optimization problem,to achieve the computational offload of edge nodes.Finally,a fault diagnosis system based on cloud edge collaborative computing is designed and implemented,and a flexible human-machine interface is added to facilitate management personnel to operate the system.This paper verifies the effectiveness and feasibility of the proposed diagnostic Bayesian network model and service composition optimization algorithm through the actual case of stacker fault diagnosis.The experimental results show that the fault diagnosis system based on cloud edge collaborative computing can improve the accuracy,real-time and energy efficiency of the diagnosis results,at the same time,it can also reduce the computing pressure on the edge nodes.