In this paper a direct adaptive neural-network control strategy for unknown nonlinear systems is presented. The system considered is described by an unknown NARMA model and a feedforward neural network is used to learn the system. Taking the neural network as a neuro model of the system, control signals are directly obtained by minimizing either the instant difference or the cumulative differences between a setpoint and the output of the neuro model. Since the training algorithm guarantees that the output of the neuro model approaches that of the actual system, it is shown that the control signals obtained can also make the real system output close to the setpoint. An application to a flow-rate control system is included to demonstrate the applicability of the proposed method and desired results are obtained.
Bibliographical noteFunding Information:
The authors gratefully acknowledge computer support from the Control Systems Centre (University of Manchester Institute of Science and Technology, U.K.).
Manuscript received July 6, 1995; revised August 11, 1997. J. R. Noriega was supported by CONACYT (México). The authors are with the Department of Paper Science, UMIST, Manchester M60 1QD, U.K. Publisher Item Identifier S 1045-9227(98)00384-1.
- Fault-tolerant control
- Flow-rate systems
- Multilayer perceptrons
- Neural networks
- Nonlinear systems