Nonlinear dynamic systems are identified and controlled using higher order multi-layer neural networks and their performance is compared with the performance of conventional multi-layer neural networks. Computer simulation results reveal that higher order neural network models are more effective in the control of systems with complex characteristics. The formation of fuzzy logic with complex input-output relationships can be replaced by neural networks with enough examples of input and output sample pairs. Computer simulation shows that the identification process using neural networks is very effective in modelling the fuzzy system whose fuzzy rule can not be obtained easily.
