Prediction of Lorenz Chaotic Time Series via Genetic Algorithm

Abstract

In this paper a method for time series prediction of chaotic systems is developed in order to increase the time horizon of prediction. Also it is assumed that the type of chaotic time series is known. In this investigation, the parameters of the chaotic system are estimated by minimizing the summation of absolute value of errors using Genetic Algorithm (GA). The results show that it is impossible to estimate accurate value of parameters because of high sensitivity of system parameters. However, it is shown that it is possible to have a model with different parameters but with similar behavior. The performance of the proposed method is investigated on Lorenz chaotic time series. The results demonstrate that the proposed method can considerably improve the horizon of prediction.

Jafar Saleh

Sr. Automation & Control Engineer

My research interests include Industrial automation, Artificial intelligence and Robotics.