Research On Ensuring Approach Of Smart Substation Communication Performance Based On Message Transmission

Smart substation is an important foundation and support of the construction of smart grid.One of the characteristics of smart substation is realizing data collection,protection,monitoring and control through communication between IEDs(Intelligent Electronic Devices)within substation.Therefore,ensuring the communication performance of smart substation is a necessary condition to the normal operation of smart substation.With the development of communication technology,A “share network” approach to the design of an smart substation communications network has been developed to connect all IEDs in the station to the same network.Although this approach can improve the efficiency of the network,it also poses challenges for the real-time performance and reliability of communication.Whether the transmission of messages in network of smart substation is reliable and in real-time is an important basis for judging the real-time and reliability of communication.At the same time,the mutual influence of message transmission behaviors is also an important factor affecting communication performance.In order to ensure the real-time and reliable transmission of normal messages in the smart substation communication network with “share network” scheme,this paper studies the guarantee method of communication performance based on analyzing the transmission behavior of messages.In order to definitize the factors that affect the communication performance of smart substation and their influencing modes,on the basis of analyzing the performance index,the structure of the communication network and the type of the communication service in the smart substation,this paper establishes the packet transmission model,obtains the calculation formula of the message transmission time,analyzes the generation mode of the waiting time and the influence of th e flow rate on the message transmission time and packet loss rate.The effects of data rate,message length and message flow rate on message transmission time and packet loss rate are analyzed by simulation model.The results show that when the network design is reasonable and the physical medium is normal,the transmission time and packet loss rate are mainly affected by flow rate.When abnormal messages with big flow rate appear in the network and causes network congestion,the transmission time and packet loss rate can not meet the requirements.In order to ensure the reliable transmission of important messages and improve the real time and reliability of communication in the case of network congestion caused by abnormal messages,it is necessary to ident ify and classify messages according to the transmission behavior then provide differentiated services.On the basis of studying Qo S technology and its application in smart substation communication network,analyzing the flow characteristics of each type of messages in the network this paper proposes a communication service strategy of smart substation based on message information label.Aiming at the network congestion caused by abnormal messages with large flow rate,this strategy implements identification and filtering of messages by condensing message source type and message type into a message information label at IEDs and checking it along with flow rate at switches.To solve the problem that weighted round robin scheduling algorithm is easily affected by message length,a scheduling method based on CLWRR?LLQ(Custom Length Weighted Round Robin with Low Latency Queue)is provided for service strategy to realize reasonable allocation of network resources.Simulation results show that this strategy can effectively identify and discard the disguised abnormal messages with big flow rate,avoid the transmission timeout and loss of important messages caused by network congestion,reduce the transmission time of other normal messages,and improve the real-time and reliability of communication.The characteristics of the data flow can reflect the transmission behavior of the message.In order to detect the abnormal messages spread in the smart substation communication network,an abnormal data flow detection metho d based on k-dist-DBSCAN and IPSO-OCSVM is proposed.This method uses k-dist-DBSCAN(Density Based Spatial Clustering of Applications with Noise)algorithm to cluster the characteristic parameter samples of the normal data flow to form the sample cluster of the normal data flow.On this basis,the IPSO-OCSVM(Improved Qarticle Swarm Optimization and One Class Support Vector Machine)algorithm are used to establish the corresponding OCSVM model according to each normal data flow sample cluster,and the abnormal detection is carried out according to these models of the normal data flow samples.Aiming at the problem that DBSCAN algorithm needs to preset neighborhood parameters,this paper provides a parameter adaptive setting method based on k-dist graph to improve DBSCAN algorithm.Aiming at the problem that OCSVM algorithm is greatly affected by error-limited parameter and kernel function parameter,an IPSO algorithm is provided to optimize the parameters of OCSVM algorithm.Simulation results show that the proposed method can adaptively build a set of normal behavior models based on the training set and improve the true negative rate of abnormal.In order to comprehensively reflect the communication status of intelligent substation and provide reference for the timely treatment,a Comprehensive evaluation method of communication performance of smart substation based on fuzzy theory is proposed.Aiming at the fuzziness of each communication status evaluation factor value in the evaluation process,a method based on fuzzy mathematics theory to deal with each communication performance factor value is proposed.This method maps the factor value to the membership degree of each evaluation grade through the membership function.According to the differences of the importance of each evaluation factor in reflecting the advantages and disadvantages of communication status and the mutual influence among the factors,a weight determination method based on the analytical hierarchy process and Spearman correlation coefficient method is proposed,which can determine the weight from subjective and objective aspects.The comprehensive evaluation method determines the evaluation result according to the weight and membership degree of each evaluation factor.Several communication cases are simulated and example analysis is given according to the simulation results.The analysis results show that this method can fully and accurately reflect th e communication status of smart substation.