A Novel Control Approach Utilizing Neural Network for Efficient Microgrid Operation with Solar PV and Energy Storage Systems

Abstract

This article introduces a novel approach for controlling a single-phase grid-connected inverter using neural network technology. While previous studies have primarily focused on voltage control techniques to facilitate power transfer in such systems, this paper advocates for the application of artificial intelligence for enhanced efficiency. Specifically, the proposed control method employs a neural network trained for function approximation to optimize power exchange between the microgrid and the main power grid. To manage battery operations, a bidirectional converter is utilized, ensuring efficient charging and discharging. During grid integration mode, voltage regulation within the microgrid is overseen by the single-phase inverter, whereas boost converters take charge during isolation mode. Results demonstrate a considerable enhancement in power management between the microgrid and the grid, alongside effective voltage regulation of the DC bus.

Keywords

• neural network
• battery energy storage system
• microgrid
• dc-dc converter
• anfis
• mppt