scholarly journals Automatic Generation of a Type-2 Fuzzy System for Time Series Forecast based on Genetic Programming

2018 ◽  
Author(s):  
Marco Antônio da Cunha Ferreira ◽  
Ricardo Tanscheit ◽  
Marley Vellasco
Keyword(s):  
Time Series ◽  
Genetic Programming ◽  
Fuzzy System ◽  
Automatic Generation ◽  
Time Series Forecast

Este trabalho descreve um Sistema Fuzzy do tipo 2 desenvolvido automaticamente com o auxílio da Programação Genética para aplicação em previsão de séries temporais. O modelo resultante, denominado GPFIS-Forecast+, é baseado no GPFIS-Forecast desenvolvido anteriormente, que fez uso da Programação Genética Multigênica com bons resultados. Os resultados demonstram que, conforme o esperado, o sistema com conjuntos fuzzy do tipo 2 melhora o desempenho, principalmente na presença de dados ruidosos.

Penerapan Logika Samar dalam Peramalan Data Runtun Waktu

2011 ◽  
Vol 3 (2) ◽  
pp. 11-15
Author(s):  
Seng Hansun
Keyword(s):  
Time Series ◽  
Fuzzy Logic ◽  
Time Series Analysis ◽  
Fuzzy Sets ◽  
Fuzzy System ◽  
Time Series Data ◽  
Fuzzy Logic System ◽  
Series Data ◽  
Series Analysis ◽  
Logic System

Recently, there are so many soft computing methods been used in time series analysis. One of these methods is fuzzy logic system. In this paper, we will try to implement fuzzy logic system to predict a non-stationary time series data. The data we use here is Mackey-Glass chaotic time series. We also use MATLAB software to predict the time series data, which have been divided into four groups of input-output pairs. These groups then will be used as the input variables of the fuzzy logic system. There are two scenarios been used in this paper, first is by using seven fuzzy sets, and second is by using fifteen fuzzy sets. The result shows that the fuzzy system with fifteen fuzzy sets give a better forecasting result than the fuzzy system with seven fuzzy sets. Index Terms—forecasting, fuzzy logic, Mackey-Glass chaotic, MATLAB, time series analysis

A contribution to the identification in fuzzy-logic control

1990 ◽  
Vol 55 (4) ◽  
pp. 951-963 ◽  
Author(s):  
Josef Vrba ◽  
Ywetta Purová
Keyword(s):  
Time Series ◽  
Fuzzy Logic ◽  
Fuzzy Logic Control ◽  
Fuzzy Logic Controller ◽  
Automatic Generation ◽  
Linguistic Variable ◽  
Fuzzy Subset ◽  
Input Output ◽  
Logic Control ◽  
Correlation Tables

A linguistic identification of a system controlled by a fuzzy-logic controller is presented. The information about the behaviour of the system, concentrated in time-series, is analyzed from the point of its description by linguistic variable and fuzzy subset as its quantifier. The partial input/output relation and its strength is expressed by a sort of correlation tables and coefficients. The principles of automatic generation of model statements are presented as well.

Sign in / Sign up

Export Citation Format

Share Document