The Research And Application Of Approaches For Bad Data Detection And Identification In Power System State Estimation

Power system state estimation is an important part of energy managementsystem. Because of some objective factors, the measure data may contain bad data.The existence of bad data may lead to the failure of the state estimation, and thencan’t get the system running status objectively. Therefore, bad data detection andidentification is of great significance to the whole power system.Currently, most practical applications for bad data detection and identificationare based on residual error method which has excellent outcomes, whereas such kindof method owns limitations like residual contamination. Hence, a method based onmeasurement correlation is proposed, in which the measurement correlation andresidual correlation are analyzed, and the correlation coefficient of variation isutilized for bad data detection and identification.Aiming at satisfying practical need of high voltage distribution networks, apractical method for bad data processing of high voltage distribution network isproposed. By analyzing the performance of the bad data, utilizing the relationshipbetween the remote communication and remote measurement data, and utilizinginterrelationship of the remote measurement data, taking the observability andradiation of the high voltage distribution network into account, a wide-range searchmethod based on nonlinear constraints is come up with. The approach employback/forward sweep method for power flow estimation data and improve geneticalgorithm for weighted estimates, thereafter processing the bad data and obtainingthe correct state of the system. The effectiveness of this method is certified bypractical application in a city grid dispatching system.Bad data processing is a core function of the intelligent dispatch arrangementmanagement. On the basis of theoretical analysis for bad data processing methods, abad data processing platform is developed, and the development language, thedevelopment platform, etc., are introduced in detail.