Pronóstico Robusto del Factor de Destrucción de Núcleos de Transformadores mediante Métodos Multivariantes de Análisis de Datos (Forecasting of the Transformer Core Destruction Factor by means of Multivariate Methods for Data Analysis)

Guillermo A. Díaz (gdiaz@iee.unsj.edu.ar), Andres A. Romero Quete (aromero@iee.unsj.edu.ar), Enrique E. Mombello (mombello@iee.unsj.edu.ar), Norma L. Furlan (nfrulan@iee.unsj.edu.ar)



This paper appears in: Revista IEEE América Latina

Publication Date: Feb. 2013
Volume: 11,   Issue: 1 
ISSN: 1548-0992


Abstract:
In designing and building of transformers, the core destruction factor is the superposition of all effects that cause a difference between the value of the core losses calculated during the design stage of the unit and the measured value after built. In this paper, two methods for forecasting the core destruction factor are proposed. One based on Mahalanobis distance and the other one based on cluster analysis. A comparison of the results obtained by conventional calculation procedure with respect to those obtained through the proposed methodologies is developed. Finally, the expected cost savings by applying the methods proposed in this article are estimated.

Index Terms:
transformer losses, forecasting, Mahalanobis distance, clustering   


Documents that cite this document
This function is not implemented yet.


[PDF Full-Text (303)]