Research On Distribution Network Topology Error Identification Method Based On PMU/SCADA Measurement Data

With the development of economy and the increasing demand for electricity,higher requirements are put forward for the security,stability and reliability of power supply.Distribution network is a link in the power system that directly interacts with users.It is the primary task to ensure the quality of power supply and the demand of users.Distribution Management System(DMS)is a software system with high intelligence level,which can better manage and control the distribution system.As the fundament of DMS,the correct network topology is vital for various advanced application functions.However,the distribution network has a large number of nodes and complex structure,and the network topology is always changing.At the same time,the distribution network lacks real-time telemetry information and telesignalization information.In most cases,the change of switching state can not be perceived in real time,even some of the switching state is unknown,which leads to the deviation between the distribution network topology obtained by the topology generator and the actual situation.Therefore,it is of great significance for the safe,stable and reliable operation of the distribution system to correctly judge the switch state and identify the error in the network topology.The application of Phasor Measurement Unit(PMU)in distribution system provides measurement data support for the identification of distribution network topology error,which is beneficial to the identification of distribution network topology error.Using PMU/SCADA measurement data,this thesis studies the identification method of distribution network topology error.Network topology error identification is often accompanied by the operation of state estimation.Therefore,the basic principle of state estimation is briefly introduced in this thesis.Two basic methods of identifying topology error are analyzed,and the principle and process of identifying topology error are emphatically studied.The differences between transmission network and distribution network are analyzed from network structure,line model,information acquisition degree,measurement configuration,etc.Combining the advantages and disadvantages of existing methods for identifying topology error,a feasible idea for identifying topology error in distribution network is proposed.Then,according to the feasible idea,a method based on branch state function for identifying distribution network topology error is proposed.On the basis of weighted least squares state estimation,state variables are augmented,branch state variables are introduced,and corresponding branch state functions are constructed to realize the approximate transformation from discrete variables to continuous variables.The influence of branch state on node self-admittance and mutual admittance is analyzed,and then the node admittance matrix of correlated branch state variables is generated,which corresponds to the variation of state variables and measurement functions.The Jacobian matrix is enlarged,the composition of measurement data in distribution network is analyzed,the mixed measurement data composed of SCADA and PMU is selected according to the actual situation,and the data fusion is carried out by using the non-linear state estimation model to improve the accuracy and redundancy of measurement data.Finally,the mathematical model of the method proposed in this thesis is established through the derivation and calculation of the formula.Finally,an example of the distribution network system with IEEE-33 node is simulated.For all possible and reasonable single and multiple topology error in three different scenarios,the method proposed in this thesis is compared with the residual method,and the influence of measurement error on the identification effect is considered.The simulation results show that the proposed method has better identification ability and fault-tolerant rate than the residual method.In view of the actual situation of insufficient and missing remote signal acquisition in distribution network,it can effectively identify single and multiple topology error.