Adaptive Optimal Tracking Control for a Flywheel Energy Storage System without the Use of a Model

This study presents an adaptive optimal tracking control method for a Flywheel Energy Storage System (FESS) using an Induction Motor (IM) without requiring an accurate system model. The control system employs a Deep Neural Network (DNN) identifier combined with an optimal controller and a Critic Neural Network (CNN) to ensure that the FESS output power follows the reference power value. The simulation results demonstrate that the proposed method achieves high accuracy and strong self-adaptation to system variations.