Investigating Efficient Probabilistic Modeling Technique for Frequency Stability Analysis of Future Power Systems

EasyChair Preprint 11725

Abstract

This paper compares two uncertainty modelling (UM) techniques to determine the accurate and efficient technique for probabilistic frequency stability assessment in large-scale power systems. The techniques are Monte Carlo (MC) and Quasi-Monte Carlo (QMC), which have been investigated in the context of their accuracy and efficiency. The performance of the UM techniques is evaluated using metrics such as the coefficient of determination (R²) and root mean square error (RMSE). By generating an extensive set of wind-speed random samples (8760 samples/simulations), both methods demonstrate remarkable accuracy, exceeding 99% when employing 1000 simulations. However, regarding efficiency, the QMC technique is more efficient than the MC technique, achieving an accuracy of over 96.5% with a considerably smaller number of generated samples and a shorter time (300 samples and in 3 minutes). In contrast, the MC technique achieves the same accuracy level (96.5%) by generating 1000 samples and requiring nearly 12 minutes for completion.

Keyphrases: Frequency stability analysis, Monte Carlo simulation, probabilistic modelling, renewable energy

@booklet{EasyChair:11725,
  author    = {Ali Hakami and Mohammed Alzubaidi and Kazi Hasan and Manoj Datta},
  title     = {Investigating Efficient Probabilistic Modeling Technique for Frequency Stability Analysis of Future Power Systems},
  howpublished = {EasyChair Preprint 11725},
  year      = {EasyChair, 2024}}