Title: Modelling, identification, implementation and MATLAB simulations of multi-input multi-output proportional integral-plus control strategies for a centrifugal chiller system

Authors: Nicolae Tudoroiu; Mohammed Zaheeruddin; Roxana-Elena Tudoroiu

Addresses: Engineering Technologies Department, John Abbott College, 21275 Lakeshore Road, Sainte-Anne-de-Bellevue, Quebec, H9X 3L9, Canada ' Faculty of Engineering and Computer Science, Department of Building, Civil and Environmental Engineering, Concordia University, 1455 De Maisonneuve Blvd. West, Montreal, QC, H3G 1M8, Canada ' Faculty of Sciences, Department of Mathematics and Informatics, University of Petrosani, 20 University Street, Petrosani, 332006, Romania

Abstract: The objective of this paper is to investigate a new design and real-time implementation approach of a predictive proportional integral-plus (PIP) closed-loop control strategy for a centrifugal chiller HVAC system. By using the analytical model of the plant, linear discrete-time multi-input multi-output (MIMO) autoregressive moving average exogenous (ARMAX) polynomial models were developed. The structure of developed ARMAX models is straightforward, and they are suitable to capture the complex dynamics of the centrifugal chiller plant. Fundamentally, these identified models use the least-squares estimation (LSE) method to evaluate the polynomials coefficients and model parameters implemented using specific tools provided by MATLAB's system identification toolbox. The new modelling approach is beneficial for simulation purposes to prove the efficiency of the proposed closed-loop control strategy, the tracking performance, and its robustness to possible changes in the load disturbance and noise level of measurement sensors.

Keywords: centrifugal chiller; MIMO ARMAX model; PI-Plus control; least-squares estimation; HVAC control systems; MISO ARMAX PI control strategy; MIMO NMSS PIP control strategy; MATLAB Simulink R2018b software package.

DOI: 10.1504/IJMIC.2020.113290

International Journal of Modelling, Identification and Control, 2020 Vol.35 No.1, pp.64 - 92

Accepted: 25 Mar 2020
Published online: 26 Feb 2021 *

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