Machine Learning-Based Solar Power Energy Forecasting

N.C. Nath, W. Sae-Tang, C. Pirak


The expanding interest in energy is one of the main motivations behind the integration of solar energy into electric grids or networks. The exact expectation of solar oriented irradiance variety can improve the nature of administration. This coordination of solar-based vitality and exact expectations can help in better arranging and distributing of energy. Discovering vitality sources to fulfil the world’s developing interest is one of the general public’s major difficulties for the coming fifty years. In this research, two machine learning techniques utilized for hourly solar power forecasting are presented. The solar power prediction model becomes robust and efficient for solar power energy forecasting once the redundant information is removed from raw data, experimental data is transformed into a settled range, the best features selection method is chosen, four different weather profiles are made based on different weather conditions and the right time series machine learning algorithm is chosen.


Solar power energy forecasting; feature selection; weather profiles; neural network; long short-term memory

Full Text:



  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Copyright © 2020. All rights reserved.
Publisher: Society of Automotive Engineers Malaysia.
eISSN: 2550-2239
ISSN: 2600-8092