Open AccessLoad frequency regulation for multi‐area power system using integral reinforcement learning
Author(s) -
Abouheaf Mohammed,
Gueaieb Wail,
Sharaf Adel
Publication year - 2019
Publication title -
iet generation, transmission and distribution
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.92
H-Index - 110
eISSN - 1751-8695
pISSN - 1751-8687
DOI - 10.1049/iet-gtd.2019.0218
Subject(s) - reinforcement learning , control theory (sociology) , computer science , electric power system , stability (learning theory) , automatic frequency control , power (physics) , adaptive control , scheme (mathematics) , control (management) , artificial neural network , mathematical optimization , mathematics , artificial intelligence , machine learning , telecommunications , physics , quantum mechanics , mathematical analysis
Active load variations in uncertain dynamical power system environments affect the energy exchange and efficiency in multi‐area power systems, which could compromise the stability of power grids. Hence, model‐free load frequency control mechanisms are needed in order to sustain proper performances under such conditions. An online model‐free adaptive control scheme based on integral reinforcement learning is proposed to regulate load frequency deviations in multi‐area power systems. This scheme takes into account the generation rate constraints of the power generation units and the optimal control decisions do not employ any knowledge about the dynamical model of the power system. This approach reformulates Bellman equation and approximates the associated solving value functions and model‐free control strategies using neural networks. The adaption mechanism uses value iteration processes to evaluate the underlying modified‐Bellman equation and model‐free control strategy in real time. The performance of the adaptive learning scheme is compared with other control methodologies using challenging validation scenarios.