Building and Environment 40 (2005) 657–669 Optimal control of building HVAC&R systems using complete simulation-based sequential quadratic programming (CSB-SQP) Jian Sun à , Agami Reddy Department of Civil, Architectural and Environment Engineering, Drexel University, 3141 Chestnut Street, Philadelphia, PA 19104, USA Received 3 February 2004; accepted 24 June 2004 Abstract This paper presents a general and systematic methodology, termed complete simulation-based sequential quadratic programming (CSB-SQP), for determining the optimal control of building HVAC&R systems. This approach allows the coupling of a detailed simulation program with an efficient optimization method, namely the sequential quadratic programming (SQP) algorithm. This approachallowstheuseofaccuratecomponentmodelsofthesystemasagainstempiricalmodelsascurrentlyused,whileproviding efficientoptimalsolutionstobedetermined.Wedevelopthemathematicalbasisofthemethodologyandapplyittoasimplecooling plant system to illustrate the accuracy, efficiency and robustness of this method. The issue of implementing such an optimization under real-time control is also discussed. r 2004 Elsevier Ltd. All rights reserved. Keywords: Optimization; HVAC&R; Simulation; SQP; Control; Cooling plants 1. Introduction Over the last three decades, there has been consider- able amount of interest in developing effective building operation strategies to achieve maximum energy sav- ings. Various optimal or near-optimal operating strate- gies, associated both with and without building thermal mass and thermal energy storage systems, have been investigated for different types of building HVAC&R systems. The control strategy to optimally operate building HVAC&R systems can be divided into two broad categories whose relationship is shown in Fig. 1 [1]: (1) Deciding on best operating mode: This involves determining the type and the number of equipment to be run (such as chiller, cooling tower, condenser water and chilled water pump, etc.) which would meet the load and comfort requirements while consuming the minimum energy. Such a scheduling problem can be viewed as an integer-programming problem with the control variables being the specific combination of equipment to be operated. An important operating mode is the sequencing of chillers, cooling towers and pumps. The sequencing defines the order and conditions associated with bringing equipment online or moving them offline. (2) Deciding on optimal set point for local-controllers: This is generally a nonlinear programming problem. The potential energy saving from optimal set point control is due to the fact that at any given time and operating mode, cooling load may be met by different combination of the control variables set points. However, only one set of control set point results in minimum energy consumption of the system. As is well known [1], there exists a tradeoff between energy consumption of different equipment in a HVAC&R system that in its wide sense consists of a primary plant and air distribution sub-systems. For example, consider the condenser water loop. Increasing the fan speed of the cooling tower ARTICLE IN PRESS www.elsevier.com/locate/buildenv 0360-1323/$-see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.buildenv.2004.08.011 à Corresponding author. Tel.: +1-215-895-2736. E-mail address: js336@drexel.edu (J. Sun).