State Estimation Of Distribution Network Based On The Data Driven Method

With the continuous development of power systems toward intelligence and automation,smart distribution network plays an increasingly important role in the power system.Recently,with the continuous deepening development of smart grids,the interaction between the distribution network and the power consumption side has increased,the operation status of the smart distribution network has become increasingly complex.The data of the distribution network presents the characteristics of large data volume and multiple data sources,and it is also mixed with low-precision and incomplete bad data.How to realize the identification and the correction of bad data and the mixed measurement of multiple sources,finally realize the accurate state estimation of the distribution network,is particularly important for the safe and reliable operation of the intelligent distribution network.Based on Supervisory Control and Data Acquisition(SCADA)and Phasor Measurement Unit(PMU),conducts research on the state estimation of the distribution network.From raw data preprocessing,data classification and correction,and multi-source data mixed measurement are used to comprehensively optimize and process distribution network measurement data to improve the accuracy of state estimation results.Based on the principle of data verification,this paper first calculates the data quality label according to the weight,and improves the Bayesian algorithm based on the required quality label result.According to the obtained Bayesian model,the raw data can be preliminarily processed and the bad data can be realized before the state estimation.Secondly,this section proposes a state estimation model based on the quadratic equation,which converts the distribution network measurement model,PMU measurement and SCADA measurement into the form of the quadratic equation.And introduce intermediate variables to uniformly linearize the data from different measurement devices,so that the measurement equations can be easily integrated together for iterative calculations to achieve mixed measurement of PMU and SCADA multi-source data.Then,based on the proposed data processing,and bad data detection and correction methods,and the quadratic linear model of PMU and SCADA hybrid measurement,a state estimation algorithm based on multi-source data hybrid measurement is proposed.Finally,the improved IEEE33-node system is used to verify the proposed method,and the results of calculation examples show the feasibility and effectiveness of the method.