Literature DB >> 26453020

Modeling and control of non-square MIMO system using relay feedback.

D Kalpana1, T Thyagarajan2, N Gokulraj3.   

Abstract

This paper proposes a systematic approach for the modeling and control of non-square MIMO systems in time domain using relay feedback. Conventionally, modeling, selection of the control configuration and controller design of non-square MIMO systems are performed using input/output information of direct loop, while the output of undesired responses that bears valuable information on interaction among the loops are not considered. However, in this paper, the undesired response obtained from relay feedback test is also taken into consideration to extract the information about the interaction between the loops. The studies are performed on an Air Path Scheme of Turbocharged Diesel Engine (APSTDE) model, which is a typical non-square MIMO system, with input and output variables being 3 and 2 respectively. From the relay test response, the generalized analytical expressions are derived and these analytical expressions are used to estimate unknown system parameters and also to evaluate interaction measures. The interaction is analyzed by using Block Relative Gain (BRG) method. The model thus identified is later used to design appropriate controller to carry out closed loop studies. Closed loop simulation studies were performed for both servo and regulatory operations. Integral of Squared Error (ISE) performance criterion is employed to quantitatively evaluate performance of the proposed scheme. The usefulness of the proposed method is demonstrated on a lab-scale Two-Tank Cylindrical Interacting System (TTCIS), which is configured as a non-square system.
Copyright © 2015 ISA. Published by Elsevier Ltd. All rights reserved.

Keywords:  Autotuning; Interaction; Modeling; Multi Input Multi Output; Non-square; Relay feedback

Year:  2015        PMID: 26453020     DOI: 10.1016/j.isatra.2015.09.012

Source DB:  PubMed          Journal:  ISA Trans        ISSN: 0019-0578            Impact factor:   5.468


  1 in total

1.  Introducing a Novel Model-Free Multivariable Adaptive Neural Network Controller for Square MIMO Systems.

Authors:  Arash Mehrafrooz; Fangpo He; Ali Lalbakhsh
Journal:  Sensors (Basel)       Date:  2022-03-08       Impact factor: 3.576

  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.