The thesis addresses development of a simple low-voltage distribution network state estimator. A goal was to create a state estimator model, based on a weighted least square algorithm, using simulated measurements of smart meters. Model’s performance was evaluated by the simulations of possible real-case scenarios.
A proper mathematical algorithm was chosen and developed using Matlab. A simple low voltage network model was used to test the state estimator algorithm. The results of simulations confirmed the robustness of the algorithm in case of random noises in the measurements, the use of bad measurements and pseudo-measurements revealed the sensitivity of algorithm in case of bad data. This bad data was used to test the limits of the state estimator model. Possibility for detection and identification of electricity theft or instrumentation malfunction was shown using simple simulation.
|
