Resumen

SANJINES TUDELA, Gimmy Nardó. Power Electric Demand in Bolivia Analysis and projection: A Neuronal Network Application. rlde [online]. 2011, n.15, pp.45-77. ISSN 2074-4706.

The objective of this research is to analyze the demand for electricity in Bolivia, based on time series forecasting. The prediction of future data from the economic perspective is important because it is used to optimize the power allocation over time under the premise that any improvement in forecast error reduction is an improvement in consumer surplus. The prognosis for its complexity is done primarily through four phase’s harmonic models, models Arima x Sarima, Arma-Garch models and finally models based on Artificial Neural Networks. The results show that the minimum error is achieved using artificial neural networks and finally concluded that the economic benefit from the implementation of artificial neural network models in predicting the electrical power demand is presented by the reduction of incremental costs generated by the forecast error.

Palabras clave : Electrical demand; Allocation Energetics; Artificial intelligence; Artificial Neuronal Networks; Modeled in Space of Frequencies; Forecast; Models Arima x Sarima; Arma - Garch Models; Estimation of Number of Harmonicas; Transformed of Fourier.

        · resumen en Español     · texto en Español     · Español ( pdf )