Paper Info
Title
Application of a Hybrid Neural Fuzzy Inference System to Forecast Solar Intensity
Abstract
This paper presents a proposal for the use of the Hybrid Fuzzy Inference System algorithm (HyFIS) as solar intensity forecast mechanism. Fuzzy Inference Systems (FIS) are used to solve regression problems in various contexts. The HyFIS is a method based on FIS with the particular advantage of combining fuzzy concepts with Artificial Neural Networks (ANN), thus optimizing the learning process. This algorithm is part of several other FIS algorithms implemented in the Fuzzy Rule-Based Systems (FRBS) package of R. The ANN algorithms and Support Vector Machine (SVM), both widely used for solving regression problems, are also used in this study to allow the comparison of results. Results show that HyFIS presents higher performance when compared to the ANN and SVM, when applied to real data of Florianopolis, Brazil, which helps to reinforce the potential of this algorithm in solving the solar intensity forecasting problems.
Year
DOI
Venue
2016
10.1109/DEXA.2016.044
2016 27th International Workshop on Database and Expert Systems Applications (DEXA)
Keywords
Field
DocType
Artificial Neural Networks, Hybrid Neural Fuzzy Inference System, Solar Forecasting, Support Vector Machines
Data mining,Neuro-fuzzy,Computer science,Support vector machine,Fuzzy logic,Artificial intelligence,Adaptive neuro fuzzy inference system,Regression problems,Artificial neural network,Machine learning,Fuzzy inference system,Fuzzy rule
Conference
ISSN
ISBN
Citations 
1529-4188
978-1-5090-3636-3
0
PageRank 
References 
Authors
0.34
7
6
Name
Order
Citations
PageRank
Silva, F.145.90
Brigida Teixeira202.37
Nuno Teixeira300.34
Tiago Pinto46825.43
Isabel Praça521240.45
Zita A. Vale639085.67