Y. Xue et J. Watton, DYNAMICS MODELING OF FLUID-POWER SYSTEMS APPLYING A GLOBAL ERROR DESCENT ALGORITHM TO A SELF-ORGANIZING RADIAL BASIS FUNCTION NETWORK, Mechatronics, 8(7), 1998, pp. 727-745
A Radial Basis Function network is a one-hidden layer feed forward typ
e network, which is special for its multidimensional centre to each hi
dden neuron. A key issue is how to set the hidden neurons and select t
heir centres to make the network efficiently converge to the required
target. A method is proposed in this study for self-organising hidden
neurons and training the centres and weights by the proposed global er
ror decent (GED) algorithm along with a modified genetic algorithm (GA
) and the method of least squares (LS). Its effectiveness is illustrat
ed using two non-linear dynamic modelling examples of a fluid power co
mponent and a system. (C) 1998 Elsevier Science Ltd. All rights reserv
ed.