This paper presents a comparative analysis of two evolved neural networks for control. Traditionally, the structure of Radial Basis Functions Networks (RBFNs) and Multilayer Feedforward Networks (MFNs) are found by a trial-and-error process. This process consists on finding an appropriate network structure such that the unknown nonlinearities of the plant can be estimated to some desired accuracy. In general, a neural network is composed of two elements: structural and learning parameters. The structural parameters are all those elements that determine the size of the network. The learning parameters are all those elements that determine learning and convergence of the network. The approach presented in this work uses a Genetic Algorithm (GA) to evolve the structure, and uses a gradient descent algorithm to adjust the weights in the network. An analysis of the evolution of RBFNs and MFNs by means of a GA is examined in detail. It is shown that the networks can be encoded in a chromosome for their evolution. Experimental results show the performance of Evolved Radial Basis Functions Networks and Evolved Multilayer Feedforward Networks in the identification and control of a nonlinear plant.
[1]
Ralf Salomon.
Improved Convergence Rate of Back-Propagation with Dynamic Adaptation of the Learning Rate
,
1990,
PPSN.
[2]
Richard D. Braatz,et al.
On the "Identification and control of dynamical systems using neural networks"
,
1997,
IEEE Trans. Neural Networks.
[3]
Lawrence Davis,et al.
Training Feedforward Neural Networks Using Genetic Algorithms
,
1989,
IJCAI.
[4]
Joos Vandewalle,et al.
Nonlinear systems identification using RBF neural networks
,
1993,
Proceedings of 1993 International Conference on Neural Networks (IJCNN-93-Nagoya, Japan).
[5]
Joos Vandewalle,et al.
Identification of nonlinear discrete-time multivariable dynamical systems by RBF neural networks
,
1994,
Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).