A new control scheme for the adaptive control of Hammerstein systems is presented. This control scheme is based on compensating the static input nonlinearity of the plant and designing a standard controller for the resulting linear system. Since the input nonlinearities of real plants are generally unknown and/or time varying, artificial neural networks (ANN) are used for the on-line identification as well as for the compensation. An LQ controller is employed for the control of the remaining linear part of the system. The results of simulation studies and real-time experiments on a hydraulic positioning system are presented to demonstrate the features and performance of the new structure.
Print ISSN: 0178-2312
Volume: 48, 11/2000
Pages: 547