Seamless Migration from Grid-Connected Photovoltaic System to Autonomous Operation Boosted by Neural Network Innovations

Department of Electrical and Electronics Engineering, Institute of Aeronautical Engineering, Hyderabad, Telangana, India

^* Corresponding author: author@email.org

Abstract

In this research introduces a method to control strategy of a solar photovoltaic (PV) system connected to the grid. For Maximum Power Point Tracking (MPPT) of the PV array a voltage source converter (VSC) is used in this system. Load at PCC receives the AC power from DC power obtained from PV array. The Positive Sequence Components (PSCs) of unbalanced grid voltages are calculated to evaluate unit templates, and an Artificial Neural Network (ANN) controller is used to reduce Total Harmonic Distortion (THD). The standalone mode is selected when there is failure in the grid reduction in grid current distortions found to be less than 5%. Additionally, a multi-variable filter-based control algorithm is employed to compensate for harmonic load currents and ensure unity power factor operation. A synchronization control ensures smooth connection to the grid when available and disconnection when unavailable. Experiments were conducted under various steady state as well as dynamic conditions. Artificial neural network (ANN) is used as extensive method which give better performance and total harmonic distortion.