Fractional order neural networks for system identification

• Published in: Chaos, solitons and fractals Vol. 130; p. 109444
• Main Authors: Zuñiga Aguilar, C.J., Gómez-Aguilar, J.F., Alvarado-Martínez, V.M., Romero-Ugalde, H.M.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2020
• Subjects: Black box modeling; Fractional calculus; Neural networks; System identification; System identification; Neural networks; Black box modeling; Fractional calculus
• ISSN: 0960-0779; 1873-2887
• DOI: 10.1016/j.chaos.2019.109444

•A new fractional order learning algorithm is proposed in this paper.•The proposed fractional order neural network (neural network trained by the proposed fractional order learning algorithm) leads to accurate and simple system identification models.•On the three different systems that were identified, the proposed fractional order neural network reaches the best accuracy with less number of parameters. Neural networks and fractional order calculus have shown to be powerful tools for system identification. In this paper we combine both approaches to propose a fractional order neural network (FONN) for system identification. The learning algorithm was generalized considering the Grünwald-Letnikov fractional derivative. This new black box modeling approach is validated by the identification of three different systems (two benchmark systems and a real system). Comparisons vs others approaches showed that the proposed FONN model reached better accuracy with less number of parameters.