×

Automated nonlinear system modelling with multiple neural networks. (English) Zbl 1260.93046

Summary: This article discusses the identification of nonlinear dynamic systems using multi-layer perceptrons (MLPs). It focuses on both structure uncertainty and parameter uncertainty, which have been widely explored in the literature of nonlinear system identification. The main contribution is that an integrated analytic framework is proposed for automated neural network structure selection, parameter identification and hysteresis network switching with guaranteed neural identification performance. First, an automated network structure selection procedure is proposed within a fixed time interval for a given network construction criterion. Then, the network parameter updating algorithm is proposed with guaranteed bounded identification error. To cope with structure uncertainty, a hysteresis strategy is proposed to enable neural identifier switching with guaranteed network performance along the switching process. Both theoretic analysis and a simulation example show the efficacy of the proposed method.

MSC:

93B30 System identification
93C10 Nonlinear systems in control theory
93B20 Minimal systems representations
93C55 Discrete-time control/observation systems
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] DOI: 10.1109/TNN.2002.1031939 · doi:10.1109/TNN.2002.1031939
[2] DOI: 10.1080/00207729708929362 · Zbl 0875.93081 · doi:10.1080/00207729708929362
[3] DOI: 10.1109/TAC.1974.1100705 · Zbl 0314.62039 · doi:10.1109/TAC.1974.1100705
[4] Bishop CM, Neural Networks for Pattern Recognition (1995)
[5] DOI: 10.1080/002071799220795 · Zbl 0938.93512 · doi:10.1080/002071799220795
[6] DOI: 10.1080/00207720701748224 · Zbl 1283.93161 · doi:10.1080/00207720701748224
[7] DOI: 10.2307/2290725 · Zbl 0776.62066 · doi:10.2307/2290725
[8] DOI: 10.1080/00207179208934272 · Zbl 0757.93015 · doi:10.1080/00207179208934272
[9] DOI: 10.1016/S0005-1098(01)00072-3 · Zbl 1112.93329 · doi:10.1016/S0005-1098(01)00072-3
[10] DOI: 10.1080/00207720600799468 · Zbl 1111.93087 · doi:10.1080/00207720600799468
[11] DOI: 10.1109/TNN.2003.820657 · doi:10.1109/TNN.2003.820657
[12] Hardle W, Applied Nonparametric Regression (1998)
[13] DOI: 10.1080/00207720802083018 · Zbl 1233.93097 · doi:10.1080/00207720802083018
[14] DOI: 10.1016/S0005-1098(01)00028-0 · Zbl 0989.93082 · doi:10.1016/S0005-1098(01)00028-0
[15] DOI: 10.1080/002071799220786 · Zbl 0938.93513 · doi:10.1080/002071799220786
[16] DOI: 10.1109/PROC.1976.10284 · doi:10.1109/PROC.1976.10284
[17] DOI: 10.1109/TNN.2002.804317 · doi:10.1109/TNN.2002.804317
[18] DOI: 10.1109/TSMCB.2003.811520 · doi:10.1109/TSMCB.2003.811520
[19] DOI: 10.1016/j.automatica.2006.03.004 · Zbl 1117.93314 · doi:10.1016/j.automatica.2006.03.004
[20] DOI: 10.1137/0111030 · Zbl 0112.10505 · doi:10.1137/0111030
[21] DOI: 10.1109/9.159571 · Zbl 0756.93043 · doi:10.1109/9.159571
[22] DOI: 10.1109/72.105416 · doi:10.1109/72.105416
[23] DOI: 10.1109/72.279191 · doi:10.1109/72.279191
[24] Werbos P, Handbook of Intelligent Control (1992)
[25] DOI: 10.1016/j.neucom.2005.12.122 · Zbl 05184843 · doi:10.1016/j.neucom.2005.12.122
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.