Application of linear regression models to determine the current and voltage harmonics. []. , 35, 3, pp. 234-241. ISSN 1815-5901.

This paper assesses the individual voltage harmonics as a function of the individual current harmonics using the statistical analyses of simple linear regression, polynomial regression and multiple linear regression. For model choice, the coefficient of determination R² and Akaike information criterion (AIC) are used. A mining process electrical system located in the Atacama Copiapó Chile, is used as a case study. This uses the technique copper electrowining as the main part of its production process. The results for different statistical models are shown and their information discussed for the power quality study. Finally, using the model that better fits to the measurements of voltage and current harmonics, some predictions for the dominant voltage harmonic are shown

: harmonics; power quality; multiple linear regression.

        · |     · |     · ( pdf )