Transmission Line Parameter Estimation Based On Measurement Data

Accurate parameters of transmission lines are critical for power system operation,control and planning.With the development of power grids,the tranditional methods to obtain model parameters such as offline test and theoretical calculation with designed layout could no longer meet the requirements for security and stability of power system.With the improvement of SCADA and WAMS data,parameter estimation based on measured data establishes the foundation for obtaining reliable transmission-line parameters.Theoretically,estimation results will be more accurate if more data points are used.But in field power systems,due to the measurement error and unknown of the measurement level,inappropriate data tend to increase the chance of nonconvergence of estimation.Therefore,it is necessary to select appropriate measured data for estimation.Fistly,the measured data with large error should be removed.What's more,according to the unique measurement accuracy of transmission lines,it is necessary to select appropriate data selection scheme for different transmission lines to obtain reliable parameters.According to the characteristic of measured data in field power systems,adaptive data selection problems in parameter estimation are studied systemically.The main research work and result are as follows:(1)In field power systems,due to the difficulty to obtain truth values of measured data,it is hard to evaluate the absolute error and accuracy level.To improve the reliability of the estimation parameters,a consistency index is defined in this paper,which can be used for removing bad data for parameters estimation.The Latin hypercube sampling technique is adopted in this chapter to numerically analyze the relationship between consistency index and measurement error.Real SCADA and WAMS data from several substations are investigated with the index and the influence factor of measurement error is analyzed.(2)On the basis of removing bad measured data,the transmission line parameter estimation model with adaptive data selection scheme is proposed considering different transmission lines have different measurement accuracy.Firstly,time window and number of data points are introduced in the estimation method as additional variables to be optimized.Since the absolute error cannot be obtained due to the unavailability of accurate transmission line parameters,relative standard deviation(RSD)instead of absolute error of estimated parameters is chosen to minimize in the proposed model.What's more,to improve convergency of the estimation,a Newton iterative technique with improving initial values is adopted to avoid iteration falling into local optimum.Lastly,by simulation analysis,the bad data removing method can be used for improving the accuracy of estimation parameters with the proposed estimation method.The influence of different measurement error on estimation results was analyzed with simulated data.(3)Based on bad data rejection with consistency index and measurement-based transmission line estimation method,application interface is designed.The application interface is based on the MATLAB and a detailed introduction is done with programing language,development environment and the method how to realize the function.The measured data from D5000 system can be read by the platform and the bad data are removed with the consistency index.The parameters of different transmission lines are estimated and the estimation results can be analyzed by the platform.Firstly,the convenience and feasiblilty of the platform is tested with measured data of three typical 500kV transmission lines from Shandong province.Secondly,to verify the effectiveness of the proposed method and the practicability of the application,500kV transmission line parameters of Shandong province are estimated and analyzed with field measured data.The influence of different estimation configuration initial parameters on estimation parameters was studied.The estimation parameters result with different objective funcitons are compared.With all the transmission lines,the difference of estimation parameters results between proposed estimation model and the estimation model with fixed data selection scheme were researched.Finally,the computational efficiency was analyzed.