Resumen

PENA ACCION, Jesús Antonio; VIEGO FELIPE, Percy Rafael; GOMEZ SARDUY, Julio Rafael  y  PADRON PADRON, Arturo Enrique. Peak load forecasting for energy management at Cienfuegos University. Universidad y Sociedad [online]. 2019, vol.11, n.4, pp.220-228.  Epub 02-Sep-2019. ISSN 2218-3620.

The peak load forecasting contributes to improve energy management. In this paper, two forecast models of peak consumption and the total daily are presented for Campus “Carlos Rafael Rodríguez” of the University of Cienfuegos. The developed models are of multiple linear regression and one of non-linear type. Non-linear model is based on an artificial neural network (ANN). The data from the readings of meters were processed and the days were classified in days of low activity and days of normal activity. The results obtained show good correlation between the model outputs and measurements. It shows model quality evaluated from the correlation coefficient R2 and the medium absolute percentage error (MAPE). It is shown that the non-linear model is better than linear one and it can be used for energy management system at University of Cienfuegos to estimate with anticipation the peak load.

Palabras clave : Energy forecasting; multiple lineal regression models; artificial neural network; forecasting models; peak load; energy consumption.

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