A fuzzy methodology to improve time series forecast of power demand in distribution systems

Author

Moraes, L.A. ; Flauzino, Rogerio A. ; Araujo, M.A. ; Batista, O.E.

Author_Institution

Dept. of Electr. & Comput. Eng., Univ. of Sao Paulo - USP, Sao Carlos, Brazil

fYear

2013

fDate

21-25 July 2013

Firstpage

1

Lastpage

5

Abstract

This paper aims to introduce a methodology for choosing the best inputs and tuning a multilayer fuzzy inference system dedicated to estimate future time series power demand values in a substation feeder. On an iteration process, older data with greater correlation with the previous forecast errors are the inputs of the fuzzy system, which has as output a future demand value. It is attempted to estimate the largest possible horizon reaching the minimum forecast error. The obtained results are satisfactory, showing that the developed methodology is capable of picking a small number of inputs to forecast with accuracy different horizons.

Keywords

fuzzy reasoning; fuzzy set theory; load forecasting; power distribution; power engineering computing; time series; distribution systems; iteration process; minimum forecast error; multilayer fuzzy inference system; power demand; substation feeder; time series forecast improvement; Correlation; Estimation; Forecasting; Fuzzy logic; Power demand; Time series analysis; Vectors; Electricity distribution; fuzzy inference systems; intelligent systems; modeling and simulation of dynamic systems; time series forecasting;

fLanguage

English

Publisher

ieee

Conference_Titel

Power and Energy Society General Meeting (PES), 2013 IEEE

Conference_Location

Vancouver, BC

ISSN

1944-9925

Type

conf

DOI

10.1109/PESMG.2013.6672491

Filename

6672491