Title :
Predicting the Mackey Glass Chaotic Time Series Using Genetic Algorithm
Author :
Farzad, Mohammad ; Tahersima, Hanif ; Khaloozadeh, Hamid
Author_Institution :
Dept. of Electr. Eng., Malek Ashtar of Tech. Univ., Tehran
Abstract :
In this paper, we investigate the accurate prediction ability of the chaotic time series. The considered chaotic system is Mackey glass time series. After inspection of Mackey glass equations, we investigate sensitivity and effect of parameters. The next step is to find unknown parameters using genetic algorithm. The result of GA shows that it is impossible to estimate accurate value of parameters. Hence, the new method has been proposed to improve the prediction by finding similar structures. At last by some examples it is shown that prediction with 100 steps ahead is possible
Keywords :
genetic algorithms; time series; Mackey glass chaotic time series; Mackey glass equation; genetic algorithm; Artificial neural networks; Chaos; Equations; Genetic algorithms; Glass; Hafnium; Inspection; Polynomials; Power system modeling; Prediction methods; Chaotic Time Series; Genetic Algorithm; horizon prediction; parameters; prediction. initial value; sensitivity; start point;
Conference_Titel :
SICE-ICASE, 2006. International Joint Conference
Conference_Location :
Busan
Print_ISBN :
89-950038-4-7
Electronic_ISBN :
89-950038-5-5
DOI :
10.1109/SICE.2006.315603
