Integrated Green Energy Solutions, Volume 2

26Day-Ahead Solar Power Forecasting Using Statistical and Machine Learning Methods

Aadyasha Patel¹ and O.V. Gnana Swathika^2*

¹ School of Electrical Engineering, Vellore Institute of Technology, Chennai, India

² Centre for Smart Grid Technologies, School of Electrical Engineering, Vellore Institute of Technology, Chennai, India

Abstract

A viable source of energy for the future is renewable sources of energy. Expansion and evolution of this field is occurring very swiftly. With progressive change in climate taking place, there is an urgent need for clean energy to replace the conventional energy sources. The repercussions of global climate change are influencing administrators to adopt clean, green, non-polluting and unlimited sources of energy. Harnessing these renewable resources will resolve the environmental concerns of many countries. Among the multiple alternative sources available, the most promising alternative to harness is solar energy. Solar energy has the power to alleviate energy security and climate change issues worldwide. The precise prediction of solar output power is a critical criterion to confirm the reliability, stability and efficiency of the photovoltaic system. The impulsive nature of the photovoltaic power generation can be overcome by forecasting the output power.

For the economic operation of a photovoltaic power plant, the forecasts are treated as references. The benefit of forecasting is that it permits time to the power plants to do the essential ...