Data analytics based neuro-fuzzy controller for diesel-photovoltaic hybrid AC microgrid

Abstract

The diesel-photovoltaic (PV) based hybrid AC microgrid systems with conventional control philosophies deliver very good performance in the grid connected mode. However, once the microgrid is isolated from the main grid the same philosophies which control the PV at its maximum power can make the microgrid unstable. In this connection, this study proposes a novel neuro-fuzzy controller to ensure the smooth transition of microgrid from grid connected mode to isolated mode, to retain the system stability even in isolated mode and to deliver the superior performance in grid connected mode as well. The considered artificial neural networks is trained with PMPP–Temp against VMPP characteristic, first of its kind. The fuzzy part of the controller derives the reference voltages subjected to the limits provided by the ANN. This study describes how well the data analytics can be utilised to retain the power system stability in emergencies. The proposed controller has been evaluated under different operating conditions and is exhibiting superior performance in achieving the desired control objectives. Results from the numerical simulations are confirmed from the experiments in real-time environment