Research On Traffic Modeling And Forecasting In Communication Network For Smart Power Distribution And Utilization Systems

Smart grid has become a universal trend in the power industry as it‘s the conception was highly regarded in all the word.It is a technology that will improve efficiency,reliability,and stability of the existing power grid by effectively managing and controlling the grid resources.At the core of the smart grid is the integration of communications networks with the power grid,where the most challenging issue is the management of a vast amount of smart grid communications traffic in the access network.To meet the challenge,this work presented a feature matching method for traffic modeling and a novel method of building a double-pattern method for traffic prediction.Firstly,the accuracy of traffic modeling and prediction has received significant interest in various domains,such as capacity planning,anomaly detection,admission control,and traffic engineering.Self-similarity and multi-fractal have attracted much attention in network traffic modeling and prediction field.And besides these two characteristics,traffic in power distribution communication network also shows a kind of significant deterministic behavior.Therefore,this thesis analyzes the notable features for traffic in power distribution and utilization communication system.Secondly,an efficient traffic model should be able to capture the most important features of the network traffic flows.And for the sake of this reason,this thesis presents an analytical study of the traffic in power distribution communication network and proposes a new feature matching model.This model divides the traffic into deterministic and chaotic flows based on MEM spectrum analysis.And for chaotic flow,according to its time scale behaviors,the L-system model of small time scale is applied into the FGN model of large time scale.The proposed model can not only capture the distribution probability faithfully but also depict the self-similarity and multi-fractal characteristics of the traffic.Finally,for the large-time scale traffic variation,it shows both a daily pattern and an hour pattern,which means the model based on single trend has not met the needs of prediction.Therefore,by dealing with the internal relationship of large-time scale network traffic,this thesis combines the regular trend and the smooth of hours and days.And then fit the double-pattern model to predict large-time scale traffic and analyze the association in it.The proposed model effectively identified the correlations of data between days and hours,and is successful in forecasting approaches.