Research And Simulation On Distribution System State Estimation And Its Application Technology

Distribution network is the link of power system that facing consumers directly. Thereliability and quality of power supply is closely related to the distribution automation level.But at this stage, the shortage of real time status information cannot reach requirements of theautomatic control functions. Therefore, obtaining the real time state of distribution system hasbecome more critical to increase the automation level. State estimation (SE) has been appliedwidely in transmission grid, and can effectively acquire system real time state. However, SEhas not been promoted in the distribution network because of the significant differencesbetween distribution and transmission network. According to the characteristic of thedistribution network, this paper does methodological and experimental research ondistribution system state estimation (DSSE) and its application technologies. The maincontent of the article is as follows:(1) The improved DSSE methods are proposed by combining the power systemmeasurement conversion technology and the weighted least squares estimation theory.Thesingle phase algorithm was extended to three phase estimation method. And the feasibilityand the estimated quality of the algorithms are verified through simulation test.(2) Considering the lack of real-time measurements, the gaussian mixture model is usedto produce pseudo-measurement with high accuracy, and improve observability of the system.In addition, the method that improve the ability to resist gross error of DSSE through acorrection method of measuring weight is presented by combining robust estimation theorywith weighted least squares theory.(3) Classification of distributed generation and the models of different kinds ofdistributed generation in DSSE are proposed. The simulation results demonstrate thefeasibility and point out that the installed distributed generation with high accuracymeasurements can reduce the error of result in the estimation.(4) The future work on improvement of the application technology of DSSE are alsodiscussed.All simulation experiments of proposed methods has been implemented in the standard distribution network model on MATLAB7.0. The methods and conclusions of this paper hascertain reference value.