Font Size: a A A

Design And Implementation Of Distribution Network Communication Simulation Subsystem

Posted on:2023-03-30Degree:MasterType:Thesis
Country:ChinaCandidate:Q HeFull Text:PDF
GTID:2542306914457784Subject:Computer technology
Abstract/Summary:PDF Full Text Request
The electricity grid is very important for the sustainable development of energy and electricity.It is an important link to the current energy supply system and in connection with China’s energy security.The development of the electricity grid is an important guarantee for the promotion of modern production and life and electrical demand of modern society.The distribution network is an important public infrastructure for national economic and social development,which plays a role in the distribution of electrical energy in the electricity grid.The distribution system has changed from the traditional to the informationphysical system.Physical system and information system are closely linked.Compared to the traditional distribution network,multi-source,heterogeneous,complex network,frequent interaction in the communication process and vulnerable to attacks.Therefore,the investigation of the distribution network communication simulation system is of great importance.This thesis designs and implements the distribution communication simulation subsystem for power grids,including network topology management,service monitoring,DDoS attack detection and network service prediction.The key to the first research on this topic is DDoS attack detection.This paper proposes a DDoS attack detection method based on information entropy and volume neural network.Under the condition of ensuring the high detection rate and low false positive rate of DDoS attack detection,the shortening of detection time is conducive to the real-time performance of DDoS attack detection.The key of the second research on this topic is the prediction of network traffic.In this thesis,a traffic prediction method based on neural graphs and long-term storage networks is proposed.The historical network traffic data is used as input of the model,and the graph network is used to capture the topology of network traffic.After receiving the spatial characteristics,the data with spatial characteristics are entered into the long-term storage network model.The long-term storage function of LSTM information is used to obtain the dynamic changes of data and the temporal properties.Finally,the spatial and temporal correlation of services in the network topology is used to predict network services.First,the research background and the significance of this work are presented and the research objectives of this work are defined.Next,the needs analysis of the research objectives is performed and the two important problems of the DDoS attack detection algorithm and the network traffic prediction algorithm in the distribution network communication subsystem are analyzed.Subsequently,the important problems that the system should solve are examined and implemented.Afterwards,the overall architecture and key functional modules of the system are designed and the design and implementation of the system is completed.Subsequently,some test examples for the typical functions and scenarios of the system are designed and the main functions of the system are tested.Finally,the article summarizes the completed work and looks forward to future work.
Keywords/Search Tags:information entropy, convolution neural network, distributed denial of service, attack graph convolution neural network, traffic prediction
PDF Full Text Request
Related items