基于数字孪生的云网智能运维技术研究OACSTPCD

Cloud-network integration is developing at an accelerated pace,which not only drives the digitalization and intelligence upgrade of communication networks,but also brings about increasing complexity in cloud-network operations and maintenance.Despite the progress made in recent years through various intelligent technologies,making network management and control more agile and efficient,large-scale cloud-network facilities still face challenges such as low efficiency,long cycles,and high costs in the operation and maintenance process.In response to these challenges,we propose three intelligent operation and maintenance techniques based on digital twins:adaptive detection,dual evaluation,and optimization adjustment,aiming to improve the efficiency of cloud-network operation and maintenance and assist in predicting network anomalies.In the adaptive detection technology,historical time series data samples are con-structed using statistical methods,and an appropriate probability distribution is selected through algorithms to predict the probability of failure occurrence.In the dual evaluation technology,both the twin system and the physical system are subject to dual evaluation to verify the causes of failures and conduct fault traceability.In the optimization adjustment technology,large-scale data is handled through tensor decomposition to optimize data samples,and machine learning is utilized to train the sample data and optimize the adjustment of intelligent operation and maintenance models.Experimental verification shows that the proposed technologies can predict network anomalies,rapidly locate faults,and optimize adjustment systems,thereby achieving the goal of intelligent operation and maintenance.