State estimation of voltage and frequency stability in solar wind integrated grids using multiple filtering techniques.

Abstract

The increasing integration of solar and wind energy into modern power grids introduces challenges in maintaining voltage and frequency stability due to their intermittent and uncertain nature. This study evaluates the performance of three advanced state observers: extended Kalman filter (EKF), unscented Kalman filter (UKF), and cubature Kalman filter (CKF) for real-time monitoring and stability assessment in solar and wind-integrated grids (SAWIG). The analysis focuses on estimation accuracy, convergence speed, and classification performance under varying phasor measurement unit (PMU) sampling rates. Simulation results reveal that the CKF achieves the lowest root mean square error (RMSE) of 0.005 at a 10 Hz sampling rate, outperforming the UKF (0.007) and EKF (0.010). In terms of dynamic performance, CKF stabilizes within 0.1 s, while UKF and EKF require 0.2 and 0.4 s, respectively. Classification evaluation shows that CKF achieves the highest accuracy of 99.5%, with precision, recall, and F1-score of 99.2, 99.3, and 99.4%, respectively. In contrast, UKF reports values of 98.8, 98.5, 98.7, and 98.6%, while EKF records 97.6, 96.9, 97.1, and 97.3%. Confusion matrix analysis further confirms a classification accuracy of 95% for CKF. These results demonstrate its robustness, speed, and precision in ensuring reliable state estimation for voltage and frequency stability in renewable-integrated smart grids.