Research On Bad Data Identification And Robust State Estimation Method Of Power System

State estimation is the core function of the energy management system of the power control center,and it is the guarantee of the basic data of advanced applications such as power system online safety analysis and dispatch control.However,there may be errors in the measurement data of power voltage and switching state as well as the parameters of the power network model.If these errors cannot be effectively identified and corrected,the state estimation results will be seriously distorted,which will directly affect the failure of decision-making of the advanced application function of the control center.Therefore,this paper studies the bad data identification and robust state estimation of power system.The topic of this paper has important theoretical significance and practical value.The main work and results of this paper are as follows:(1)An identification method of complex bad data combining power flow tracking and Lagrange multiplier is proposedIn order to solve the problem that it is difficult to accurately identify a variety of bad data in a single section,this paper proposes a complex bad data identification method combining power flow tracing and Lagrange multiplier method.Based on the principle of node power balance and the consistency principle of the head and end power flow of the branch,and the classification and prediction heuristic rules of bad data,the detection and correction of typical bad data and the detection of suspicious bad data are realized through the forward-backward flow tracking.The identification accuracy of bad data is further improved by detecting and correcting the bad data in suspicious bad data by Lagrange multiplier method of bad data identification.The result of the experiment shows that the proposed method has good identification effect on complex bad data in single section.(2)A forward-backward robust state estimation method of radiation network with bad data preprocessing is proposedIn order to solve the problem of weak robustness of existing robust state estimation methods for high proportion and strong correlation data,a forward-backward robust state estimation method of radiation network with bad data preprocessing is proposed in this paper.On the basis of single branch and single node state estimation,the robust performance of forward backward state estimation is effectively improved through the measurement classification normalized exponential weight function strategy and forward backward full range estimation strategy.Moreover,the heuristic rules of typical bad data detection and correction are embedded in the forward backward process,which further improves the robust performance of state estimation.The result of the experiment shows that the proposed method has good robust performance when there are high proportion of strong correlation bad measurements and bad measurements and parameter errors simultaneously.(3)A forward-backward robust state estimation method for simple ring network with bad measurement preprocessing is proposedIn order to solve the problem that residual pollution and inundation are more serious in the ring network,a forward-backward robust state estimation method for simple ring network with bad measurement preprocessing is proposed in this paper.Through topology analysis,the complex ring network is decomposed into a radial network subsystem and a single ring network subsystem.Then,through the minimum unbalanced power selection strategy of the unbalance point and the unloop compensation strategy based on the loop power estimation of the ring network,the equivalent unbalance of the single ring network is realized.Finally,the state estimation of complex power grid is realized based on the forward backward robust state estimation method of radial power grid and the compensation iterative correction of loop power.The result of the experiment shows that the proposed method has better robustness when there are multiple bad measurements in the ring network.