Services on Demand
Journal
Article
Indicators
Cited by SciELO
Access statistics
Related links
Similars in SciELO
Share
Ingeniería, investigación y tecnología
On-line version ISSN 2594-0732Print version ISSN 1405-7743
Abstract
IBARGUENGOYTIA-GONZALEZ, Pablo Héctor; REYES-BALLESTEROS, Alberto; BORUNDA-PACHECO, Mónica and GARCIA-LOPEZ, Uriel Alejandro. Wind power forecasting using Artificial Intelligence tools. Ing. invest. y tecnol. [online]. 2018, vol.19, n.4, e033. ISSN 2594-0732. https://doi.org/10.22201/fi.25940732e.2018.19n4.033.
There is a worldwide trend of using clean energy instead of fossil fuels, given the steady increase in energy demand and the interest in preserving the environment. Wind energy is the renewable energy which has grown the most at global level in recent years. However, there are still some difficulties to extend its use elsewhere in the country. One of these challenges is the difficulty of knowing in advance how much energy will be available to inject into the grid. This paper describes the development of a technique of Artificial Intelligence (AI) for wind power forecast using weather information for several years. Specifically, a detailed research on the potential use of Bayesian Networks for these forecasting applications was made. The need to consider the time in the forecasting process was identified and a new proposal of Dynamic Bayesian Networks (DBN) was performed. The forecast system with Dynamic Bayesian Networks was tested with data from the regional center for wind technology (CERTE) of the National Institute of Electricity and Clean Energies (INEEL) in Oaxaca, Mexico. Our results were satisfactorily compared with forecasting results from time series techniques indicating that DBN is a promising tool for wind power forecasting.
Keywords : Wind energy; power forecast; Artificial Intelligence; Dynamic Bayesian Networks.