Regresar

Web platform for wind pbkp_rediction based on artificial intelligence techniques with the use of the arima method: Case study

Abstract:

The purpose of the following article is to develop an interactive web application to make a forecast with models that work on time series. The study carried out corresponds to an exploratory investigation, with descriptive and explanatory elements, which include the analysis of historical data and the behavior of wind speed, through unsupervised learning techniques. For the analysis of the information and the pbkp_rediction, R. was used for the development of the web application Shiny of the r-cran repository that facilitates the creation of interactive web applications directly from R. The results obtained indicated that the use of ARIMA models for the forecast on time series is acceptable reaching an AIC of -130328.1 and a BIC of -130278. In addition, it was possible to conclude that this type of analysis with ARIMA models is a good starting point to make more complex and reliable pbkp_redictive models.