INVERSION CONTROL OF NONLINEAR-SYSTEMS WITH NEURAL-NETWORK MODELING

A method of controlling certain types of nonlinear dynamical systems w hose dynamics can be modelled by a multilayer neural network is propos ed. The control algorithm assumes that the plant equations are not kno wn but the dimension of the system is known, The control input is deri ved by inversion of a forward neural network via the Newton Raphson me thod. During inversion of the multilayer neural network some optimal c ontrol senses are resolved. To suppress the control error due to the m odelling error of the forward neural network, the inversion controller with a conventional feedback controller is proposed, which provides a better performance than a pure inversion controller. The proposed alg orithm shows various advantages, and computer experiments on a bioreac tor prove the effectiveness of this algorithm.