Title :
A STLF in distribution systems - A short comparative study between ANFIS Neuro-Fuzzy and ANN approaches - part I
Author :
Santos, P.J. ; Rafael, S. ; Lobato, P. ; Pires, A.J.
Author_Institution :
Dept. of Electr. Eng. at EST Setubal, Polytech. Inst. of Setubal, Setubal
Abstract :
The STLF algorithms belong to the set of methodologies which aim to furnish more effectiveness in planning, operation and conduction in electric energy systems. Actions like, maintenance issues, network management, and eventual power purchase decisions within liberalized electricity markets require, among others, reliable next-hour load forecasts. Regressive methods are widely used. Artificial neural networks (ANN), models based on fuzzy inference (ANFIS) and neuro-fuzzy (NF) are used in short-term problems (one hour ahead). In this paper it´s made a short comparative study in order to compare these three approaches for the same case study. These methodological approaches are discussed in a real life case study.
Keywords :
fuzzy reasoning; load forecasting; neural nets; power distribution planning; power engineering computing; power markets; artificial neural networks; distribution systems; electric energy system planning; fuzzy inference; liberalized electricity markets; load forecasting; network management; neuro-fuzzy approach; power purchase decisions; Artificial neural networks; Electricity supply industry; Energy management; Fuzzy neural networks; Load forecasting; Maintenance; Power system management; Power system modeling; Power system planning; Power system reliability; ANFIS; Artificial neural network; Consumption trend; Electrical distribution network; Short-term load forecast;
Conference_Titel :
Power Engineering, Energy and Electrical Drives, 2009. POWERENG '09. International Conference on
Conference_Location :
Lisbon
Print_ISBN :
978-1-4244-4611-7
Electronic_ISBN :
978-1-4244-2291-3
DOI :
10.1109/POWERENG.2009.4915196
