TY - GEN
T1 - Artificial Intelligence Based Modelling for Energy Output Predictions of Renewables
AU - Domenikos, George Rafael
AU - Hajimirza, Shima
AU - Acar, Gizem
N1 - Publisher Copyright: Copyright © 2023 by ASME.
PY - 2023
Y1 - 2023
N2 - The reliance of energy grids to a variety of energy sources leads to complex systems that need to consider multiple parameters in order to optimize the use of their resources as to have the least environmental impact. In this work the authors present a modelling based on statistical analysis and artificial intelligence that aims to predict the energy outcome of renewable energy sources, more specifically wind and solar farms. In the presented paper traditional mathematical/ statistical approaches (like time series through interpolation) are used in combination to Artificial Intelligence algorithms. Thus, by combining the use of Artificial Neural Networks (ANN) with mathematical prediction models it is showcased that energy output of these renewable sources can be predicted with great accuracy. In addition, using a combined quasi-AI algorithm consisting of an optimized use of both mathematical and AI models is shown to lead to quickly predicting results for short term energy production with much less computational intensity compared to full AI algorithms, requiring more intensive training. Overall, using the presented models an energy grid can be expected to utilize the potential of its renewable energy sources to the fullest, minimizing dependence on non-renewables and as such the environmental damage they cause.
AB - The reliance of energy grids to a variety of energy sources leads to complex systems that need to consider multiple parameters in order to optimize the use of their resources as to have the least environmental impact. In this work the authors present a modelling based on statistical analysis and artificial intelligence that aims to predict the energy outcome of renewable energy sources, more specifically wind and solar farms. In the presented paper traditional mathematical/ statistical approaches (like time series through interpolation) are used in combination to Artificial Intelligence algorithms. Thus, by combining the use of Artificial Neural Networks (ANN) with mathematical prediction models it is showcased that energy output of these renewable sources can be predicted with great accuracy. In addition, using a combined quasi-AI algorithm consisting of an optimized use of both mathematical and AI models is shown to lead to quickly predicting results for short term energy production with much less computational intensity compared to full AI algorithms, requiring more intensive training. Overall, using the presented models an energy grid can be expected to utilize the potential of its renewable energy sources to the fullest, minimizing dependence on non-renewables and as such the environmental damage they cause.
UR - http://www.scopus.com/inward/record.url?scp=85185542334&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85185542334&partnerID=8YFLogxK
U2 - 10.1115/IMECE2023-112607
DO - 10.1115/IMECE2023-112607
M3 - Conference contribution
T3 - ASME International Mechanical Engineering Congress and Exposition, Proceedings (IMECE)
BT - Energy
PB - American Society of Mechanical Engineers (ASME)
T2 - ASME 2023 International Mechanical Engineering Congress and Exposition, IMECE 2023
Y2 - 29 October 2023 through 2 November 2023
ER -
