Time-Delay Wavelet Neural Networks Model with Application to Ship Control
A time-delay wavelet neural network (TDWNN) is introduced to realize system identification based on model of nonlinear auto-regressive with exogenous inputs (NARX). The method incorporates the delayed massage and the gradient message of the system, is able to reflect the changing tendency of system. Multi-objective optimization method is used to estimate the network coefficient. The wavelet-network-based identification model is used for online system identification, and the experiment result of ship course control proved the efficiency of the identification model.
KeywordsWavelet neural network system identification time delay
Unable to display preview. Download preview PDF.
- 1.Xiu, Z.H., Zhang, J.S., Zheng, Y.L.: Bionics in Computational Intelligence. Science Press, Beijing (2003)Google Scholar
- 3.Honik, K.: Approximation capabilities of multilayer feedforwad networks neural. Neural Network 4, 551–557 (1991)Google Scholar
- 4.Yuan, X.M., Li, H.Y., Liu, S.K., et al.: The Application of Neural Network and Genetic Algorithm in the Area of Water Science. China Water Power Press, Beijing (2002)Google Scholar
- 5.Yin, J.C., Wang, L.D., Wang, N.N.: A Variable-Structure Gradient RBF Network with its Application to Predictive Ship Motion Control. Asian Journal of Control, doi:10.1002/asjc.343 (in press)Google Scholar
- 7.Chen, J.N.: Basis of Wavelet Analysis. Shanghai University Press, Shanghai (2002)Google Scholar
- 8.Guo, Y.J., Zhang, S.C.: Wavelet Neural Network Estimation Model for Mine Safety. Journal of Northeastern University (Natural Science) 27, 702–705 (2006)Google Scholar
- 11.Gao, S.X., Cao, S.F., Zhang, Y.: Research On PID Control Based on BP Neural Network and Its Application. In: Proc. 2010 2nd International Asia Conference on Informatics in Control, Automation and Robotics, pp. 91–94 (2010)Google Scholar
- 12.Jia, X.L., Yang, Y.S.: Mathematic Model of Ship Motion. Dalian Maritime University Press, Dalian (1998)Google Scholar