Predicting Natural and Chaotic Time Series with a Swarm-Optimized Neural Network

Received Date: May 30, 2011

Published Date: October 31, 2011

Abstract

Natural and chaotic time series are predicted using an artificial neural network (ANN) based on particle swarm optimization (PSO). Firstly, the hybrid ANN+PSO algorithm is applied on Mackey–Glass series in the short-term prediction x(t+6), using the current value x(t) and the past values: x(t−6), x(t−12), x(t−18). Then, this method is applied on solar radiation data using the values of the past years: x(t−1), ..., x(t−4). The results show that the ANN+PSO method is a very powerful tool for making predictions of natural and chaotic time series.

FullText(HTML)

[1]	Han M and Wang Y 2009 Exp. Syst. Appl. 36 1280
[2]	Bezruchko B P et al 2001 Phys. Rev. E 64 056216
[3]	Farmer J D and Sidorowich J J 1987 Phys. Rev. Lett. 59 845
[4]	Zhang J S and Xiao X C 2000 Chin. Phys. Lett. 17 88
[5]	Wong F S 1991 Neurocomputing 2 147
[6]	Chng E S et al 1996 IEEE Trans. Neural Networks 7 190
[7]	Kim D and Kim C 1997 IEEE Trans. Fuzzy Systems 5 523
[8]	Jang J S R et al 1997 Neuro-fuzzy and Soft Computing (New Jersey: Prentice-Hall)
[9]	Lin D, Wang X 2011 Neurocomputing 74 2241
[10]	Fu Y Y et al 2010 Exp. Syst. Appl. 37 4441
[11]	Cho K B and Wang B H 1996 Fuzzy Sets Syst. 83 325
[12]	Chen Y et al 2006 Neurocomputing 69 449
[13]	Geethanjali M et al 2008 Neurocomputing 71 904
[14]	Lazzús J A 2010 Comput. Math. Appl. 60 2260
[15]	Kennedy J and Eberhart R 1995 IEEE Int. Conf. Neural Networks 4 1942
[16]	Lazzús J A 2009 Int. J. Thermophys. 30 883
[17]	Mackey M C and Glass L 1977 Science 197 287
[18]	Farmer J D 1982 Physica D 4 336
[19]	Mohandes M et al 1998 Renew. Energy 14 179
[20]	Sözen A et al 2004 Energy Convers. Manag. 45 3033
[21]	Rehman S and Mohandes M 2008 Energy Policy 36 571
[22]	Lazzús J A et al 2011 Appl. Solar Energy 47 66

Export File