State Estimation of IEEE 5-Bus and 14-Bus Power System

Abstract

This paper investigates techniques for precisely estimating the state (angles and voltages) of electrical networks using the 5-bus and 14-bus test systems. Two methods were examined: Weighted Least Squares (WLS), a statistical technique used to minimize measurement errors; and the Artificial Neural Network (ANN), a machine learning model trained to estimate the system states. The results showed that WLS provided reliable estimates with fewer measurements, while ANN yielded the most accurate results overall. This study contributes to the enhancement of real-time power grid monitoring, thereby supporting a stable and consistent electricity supply.