Temperature Control Of A Water Bath Process: A Comparative Study Between Neuro-Control, A Self-Tuning Adaptive Control, A Generalised Predictive Control, And A Conventional Feedback Control Approach

Currently, neural networks are being used to solve problem related to control. One way to determine the reliability of the neuro-control technique is to test it on a variety of realistic problems, and to compare directly with existing traditional control technique, to see whether it works well and where it needs further refinement. In this article, we compare the neuro-control approach to a self-tuning adaptive control approach, a generalised predictive control approach, and a conventional feedback control approach on a real-time process control system. The neuro-control scheme consists of a backpropagation through time utility where two neural networks are trained one as an emulator, and the other as a controller. The four systems are compared conceptually and through experimental studies on the same single-input single-output water bath temperature control process. Comparisons, where applicable, are made with respect to methodology, system tracking performance, speed of adaptation, disturbance rejection, effect of long time-delay, and noise rejection. The results show that the neural network controller performs very well and offers encouraging advantages in many aspects over the other three controllers.

[1]  Rubiyah Yusof,et al.  Application of generalised predictive control to a temperature control process , 1991, Proceedings IECON '91: 1991 International Conference on Industrial Electronics, Control and Instrumentation.

[2]  David W. Clarke,et al.  Generalized Predictive Control - Part II Extensions and interpretations , 1987, Autom..

[3]  Michael I. Jordan,et al.  Forward Models: Supervised Learning with a Distal Teacher , 1992, Cogn. Sci..

[4]  Kumpati S. Narendra,et al.  Identification and control of dynamical systems using neural networks , 1990, IEEE Trans. Neural Networks.

[5]  Sigeru Omatu,et al.  A neural network based control scheme with an adaptive neural model reference structure , 1991, [Proceedings] 1991 IEEE International Joint Conference on Neural Networks.

[6]  B. Widrow,et al.  Neural networks for self-learning control systems , 1990, IEEE Control Systems Magazine.

[7]  L.G. Kraft,et al.  A comparison between CMAC neural network control and two traditional adaptive control systems , 1990, IEEE Control Systems Magazine.

[8]  David Clarke,et al.  Self-tuning control , 1979 .

[9]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[10]  Paul J. Werbos,et al.  Backpropagation Through Time: What It Does and How to Do It , 1990, Proc. IEEE.

[11]  Hideaki Sakai,et al.  A nonlinear regulator design in the presence of system uncertainties using multilayered neural network , 1991, IEEE Trans. Neural Networks.