Neural networks and hydraulic control—from simple to complex applications

Abstract The control of hydraulic servo-systems has been the focus of intense research over the past decades. The highly non-linear behaviour of these devices makes them ideal subjects for applying different types of sophisticated controllers. This paper considers the application of neural networks to the control of hydraulic servo-systems subjected to either non-linear friction or highly coupled loads. Three applications are considered: a very simple quasi-open-loop pattern follower, a proportional-integral-derivative (PID) multiple-gain neural controller and a neural-based controller for a very complex multiple-input multiple-output system. A major factor in these applications is the fact that the neural network controllers have been applied to real-time control of experimental systems. In all cases, these controllers showed superior performance over conventional-type PID controllers.