Title: Guidance system for control periods under modelling threshold

Authors: Daniel Viúdez-Moreiras; Juan M. Martín-Sánchez

Addresses: ADEX SL., Spain ' IEEC Department, UNED, Madrid, Spain

Abstract: This paper presents a new guidance system for process control that can improve the control performance of model-based advanced controllers or optimised adaptive controllers, when their control periods have to be selected within a lower range of time values that may increase the sensitivity of the predictive model parameters to identification errors. The guidance system generates at each control instant a future desired trajectory for the process output variable under control, from which it computes a set point for the derivative variable of the process output variable. A predictive or adaptive predictive controller drives the derivative variable to its set point and, as a consequence, the process output converges towards said desired trajectory. The new guidance system is illustrated by its application to the level control of a MIMO nonlinear process, in the presence of noises and disturbances. The results show how the control performance of the guidance system is not affected by 'small' control periods that would deteriorate the performance of an adaptive predictive controller for the same application.

Keywords: optimised adaptive control; predictive control; adaptive predictive control; derivative control; guidance systems; guidance strategy; process control; modelling threshold; model-based control; MIMO nonlinear processes; control performance.

DOI: 10.1504/IJAAC.2016.077587

International Journal of Automation and Control, 2016 Vol.10 No.3, pp.215 - 233

Received: 03 Jun 2015
Accepted: 26 Jan 2016
Published online: 06 Jul 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article