Design and Implementation of a Neural Network Controlled Electro-Hydraulic Drive

Abstract

Hydraulic systems are commonly used in industry when large forces of torques are required. Increasing demands for positional, force and speed control from users have led to the use of closed-loop control techniques. While classical controllers are used successfully for many hydraulic applications, they cannot cope with the non-linearities inherent in hydraulic systems or the changes to system parameters over time. This paper considers these problems and proposes a solution in the form of a neural network-based controller in the forward path of the system directly controlling the hydraulic plant. A network-learning mechanism is also proposed to train the network to minimize the system error and examples are given of the implementation with comparisons against a PID controller. The examples illustrate the rapid convergence of the training algorithm and the robustness properties of a simple two-neuron, two-input network controlling a non-linear time-varying plant.

Get full access to this article

View all access options for this article.

References

Walters

Hydraulic and electro-hydraulic servo systems, 1967, Illfe Books, London).

Whatton

Fluid power systems, 1989, Prentice Hall, London).

Astrom

K. J.

Bohlin

Numerical identification of linear dynamical systems from normal operating records. In Theory of adaptive control systems, 1966, Plenum Press, New York).

Daley

A study of a fast self-tuning algorithm. Proc. Instn Mech. Engrs, Part C., 1986, 200, 425–430.

Wellstead

P. E.

Zarrop

M. B.

Self-tuning systems, 1991, Wiley, New York).

Daley

Application of a fast self-tuning control algorithm to a hydraulic test rig. Proc. Instn Mech. Engrs, Part C., 1987, 201, 285–295.

Newton

D. A.

Design and implementation of an electro-hydraulic rotary drive system, IMechE Seminar, Machines actuators and controls, 1993, pp. 25–35.

Edge

K. A.

Figueredo

K. R. A.

An adaptively controlled electro-hydraulic servo system. Proc. Instn Mech. Engrs, Part B, 1987, 201, 175–189.

Glover

Robust-stabilisation of linear multivariable systems: Relations to approximation. Int. J. Control. 1986, 43, 741–766.

10.

Maciejowski

J. M.

Multivariable feedback design, 1989, Addison Wesley, Reading, MA).

11.

Stoten

D. P.

Hodgson

S. P.

Comparative implementation studies of the minimal controller synthesis algorithms on a class 1 manipulator. Proc. Instn Mech. Engrs, Part I, 1991, 4, 263–273.

12.

Hecht-Nielsen

Neurocomputing, 1992, Addison Wesley, Reading, MA).

13.

Grossberg

Neural networks and natural intelligence, 1988, MIT Press, Cambridge, MA).

14.

Kohonen

An introduction to neural computing. Neural Networks, 1988, 1, 3–16.

15.

Le Cun

A theoretical framework for backpropagation, Technical Report CRG-TR-88–6, 1988, University of Toronto.

16.

Pinenda

F. J.

Recurrent backpropagation and the dynamical approach to adaptive neural computing. Neural Computation, 1988, 1, 161–172.

17.

Pham

D. T.

S. J.

A recurrent backpropagation network for dynamic system identification, J. Syst. Engng, 1992, 2, 213–223.

18.

Gupta

M. M.

Rao

D. H.

Nikiforouk

P. N.

Neuro-controller with dynamic learning and adaption. J. Intelligent Robotic Syst., 1993, 7, 151–173.

19.

Narenda

K. S.

Adaptive control of dynamic systems using neural networks. Handbook of intelligent control, 1992, pp. 141–183 (Addison-Wesley, Reading, MA).

20.

Chen

F. C.

Back-propagation neural networks for non-linear self-tuning adaptive control. IEEE Control Syst. Mag., 1990, 4, 45–48.

21.

Lane

S. H.

Handelman

D. A.

Gelfand

J. J.

Higher order CMAC neural networks—theory and practice. Proceedings of the American Control Conference, Boston, 1991, pp. 1579–1585 (IEEE).

22.

Kolomogorov

A. N.

On the representation of continuous functions of many variables by superposition of continuous of one function and addition. Dokl. Akad, 1957, 114, 953–956.