Feedforward artificial neural network to improve model predictive control in biological processes Online publication date: Fri, 17-Apr-2015
by Senthil Kumar Arumugasamy; Zainal Ahmad
International Journal of Automation and Control (IJAAC), Vol. 5, No. 4, 2011
Abstract: Artificial neural networks (ANNs) offer the versatility of being able to model the dynamics of a biosystem without requiring a phenomenological model. In addition, model predictive control (MPC) is a member of advanced discrete-time process control algorithms. The recent developments in the biotechnology due to MPC utilising the capability of ANN make the practical application of non-linear process control strategies a reality. This paper reviews the recent enhancement and applications of MPC in various biochemical processes using feedforward artificial neural networks which is also known as neural predictive control. The capability of neural predictive control to handle the common problems associated with biochemical processes, namely optimisation of objective function, optimisation of dynamic behaviour of the system, control of ill-defined non-linear systems, improving the computational efficiency of the strategy, disturbance rejection ability and evaluating the control effectiveness are discussed. The review clearly indicates that enormous work has been carried out involving dynamic behaviour of the bioreactor system which is analysed and optimised revealing that feedforward neural network has evolved as a good bioreactor neuro-controller.
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Automation and Control (IJAAC):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email subs@inderscience.com