Neuro-models for discharge air temperature system M. Zaheer-uddin * , N. Tudoroiu Centre for Building Studies, Department of Building, Civil and Environmental Engineering, Concordia University, Montreal H3G 1M8, Canada Received 3 March 2003; received in revised form 22 July 2003; accepted 2 August 2003 Abstract Nonlinear neuro-models for a discharge air temperature (DAT) system are developed. Experimental data gathered in a heating ventilating and air conditioning (HVAC) test facility is used to develop multi- input multi-output (MIMO) and single-input single-output (SISO) neruo-models. Several different network architectures were explored to build the models. Results show that a three layer second order neural network structure is necessary to achieve good accuracy of the predictions. Results from the developed models are compared, and some observations on sensitivity and standard deviation errors are presented. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: DAT system; HVAC systems; MIMO models; Neuro-models; SISO models 1. Introduction The discharge air temperature (DAT) control loop is one of the most important control loops in heating ventilating and air conditioning (HVAC) systems. In view of its importance, several studies have been conducted on the modeling and control of DAT systems. From the view point of modeling, dynamic analysis of cooling coils has been studied by Gartner [1], Khan [2] and Maxwell et al. [3]. Also, simplified methods, such as SISO models with delay, have been used to describe the dynamics of DAT systems [5,6]. In this study, we are interested in exploring the use of neural networks to capture the dynamics of the DAT system. We use the control oriented ap- proach to modeling of the DAT system, since a future direction of the present research is to develop intelligent control strategies for DAT systems. Neuro-models have advantages over conventional online identification methods in terms of robustness and accuracy of model predictions. The neuro-models exhibit good accuracy and Energy Conversion and Management 45 (2004) 901–910 www.elsevier.com/locate/enconman * Corresponding author. Tel.: +1-514-848-3200; fax: +1-514-848-7965. E-mail address: zaheer@cbs-engr.concordia.ca (M. Zaheer-uddin). 0196-8904/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.enconman.2003.08.004