A power control strategy for flywheel doubly-fed induction machine storage system using artificial neural network

Abstract

A large-capacity low-speed flywheel energy storage system (FESS) based on a doubly-fed induction machine (DFIM) consists of a wound-rotor induction machine and a back-to-back converter rated at 30-35% of the machine power rating used for rotor excitation. This system has been promoted as a viable mean of energy storage for power system applications as grid frequency support/control, uninterruptible power supply (UPS), power conditioning, and voltage sag mitigation. This paper presents a simple power control strategy based on artificial neural networks (ANN) to charge/discharge a flywheel DFIM (FW-DFIM) storage system while maintaining controllable grid side power. The proposed controller is based on conventional vector control system supplemented by an ANN-based current decoupling network used to develop the required rotor current components based on the required grid power level and flywheel instantaneous speed. The controller is designed to avoid overloading both stator and rotor circuits while the flywheel is charged/discharged. Additionally, it avoids using the required outer power loop or a hysteresis power controller, hence, simplifies the overall control algorithm. The validity of the developed concept along with the effectiveness and viability of the control strategy in power system applications is confirmed by computer simulation using Matlab/Simulink for a medium voltage 1000hp FW-DFIM. The simulation study is carried out for uninterruptible power supply (UPS) applications and power leveling to improve the quality of electric power delivered by wind generators.