Theory and applications of HVAC control systems – A review of model predictive control (MPC)
Introduction
With the significant increase of energy consumption in buildings, energy saving strategies have become a priority in energy policies in many countries. For instance, building energy consumption in the EU was 37% of the final energy totals in 2004 [1]. In the USA, building energy consumption accounted for 41% of primary energy consumption in 2010 [2]. The categories of building services and heating, ventilation, and air conditioning (HVAC) systems make up the major sources of energy use in buildings (almost 50% [1], [2]). Therefore, the development and implementation of effective control techniques for HVAC systems is of primary importance. In particular, with the decreased costs of data processing, storage, and communication over recent years, the design and implementation of more complex control techniques have become feasible.
Despite the similarity of HVAC control to other types of process control, certain features exist that render HVAC system control unique and challenging, including the following:
- •
Nonlinear dynamics;
- •
Time-varying system dynamics and set-points;
- •
Time-varying disturbances;
- •
Poor data due to low resolution of analog-to-digital converter (ADC) devices, sampling rates, accuracy of sensors, and lack of access to network forecasting and environmental information;
- •
Interacting and at times conflicting control loops; and
- •
Lack of supervisory control (in many buildings).
Many control methods have been developed or proposed for HVAC systems. However, because of their simplicity, on/off and PID control are still used in many HVAC systems, resulting in inconsistent performance among such systems. With advances in data storage, computing, and communication devices, it is now feasible to adopt and implement a proper control approach to overcome the inherent issues in HVAC control. The focus of this paper is on a survey of control methods for HVAC systems, and emphasis is placed on the model predictive control (MPC) approach because research on MPC development for HVAC systems has intensified over the last years due to its many inherent advantages, which include
- •
Use of a system model for anticipatory control actions rather than corrective control;
- •
Integration of a disturbance model for disturbance rejection;
- •
Ability to handle constraints and uncertainties;
- •
Ability to handle time-varying system dynamics and a wide range of operating conditions;
- •
Ability to cope with slow-moving processes with time delays;
- •
Integration of energy conservation strategies in the controller formulation;
- •
Use of a cost function for achievement of multiple objectives;
- •
Use of advanced optimization algorithms for computation of control vectors;
- •
Ability to control the system at both the supervisory and local loop levels.
However, a comprehensive survey of MPC approaches for HVAC systems is still lacking. In particular, selected trends and issues related to MPC design must be identified.
The organization of this paper is as follows. First, a review of HVAC systems is presented to outline the spectrum of control tasks in HVAC systems. Section 2 includes a brief review of previous surveys related to HVAC control. Section 3 classifies the approaches to HVAC control according to methodology, scope and implementation to create a framework with which to compare MPC with other methods. Section 4 discusses the comparison of MPC with other methods as well as the factors that affect its performance. Finally, Section 5 includes a summary of important factors that govern MPC design and outlines open design problems for HVAC systems.
Section snippets
Previous surveys
A large body of literature has been published on applications of MPC to HVAC systems, but to the best of the authors' knowledge, no recent comprehensive review has been published on the theory and applications of MPC.
Brief reviews of hard and soft control techniques were reported in Refs. [3], [4], respectively. The hard control techniques reviewed in Ref. [3] include gain scheduling, optimal control, robust control, MPC, and nonlinear and adaptive control. The soft or intelligent control
Classification of HVAC control methods
A classification for control methods in HVAC systems is illustrated in Fig. 1. The control methods are divided into classical control, hard control, soft control, hybrid control, and other control techniques. Brief details of each method are provided in the following sections.
Model predictive control (MPC)
Because the focus of this paper is MPC, a comprehensive review of MPC techniques and comparisons with other control techniques are provided in this section.
The MPC uses a system model to predict the future states of the system and generates a control vector that minimizes a certain cost function over the prediction horizon in the presence of disturbances and constraints. The first element of the computed control vector at any sampling instant is applied to the system input, and the remainder is
Conclusions
Certain important points of MPC development for HVAC control can be summarized as follows:
- •
Many attractive choices are available for HVAC system control in the form of conventional controllers, hard controllers, soft controllers, and hybrid controllers. These techniques were reviewed, and the advantages and disadvantages of each technique were highlighted. Compared with most of the other control techniques, MPC generally provides superior performance in terms of lower energy consumption, better
Acknowledgments
This research was financially supported by Ryerson Center for Urban Energy (CUE), Toronto Hydro and Mitacs-Accelerate Program.
References (132)
- et al.
A review on buildings energy consumption information
Energy Build
(2008) - et al.
Advanced control systems engineering for energy and comfort management in a building environment – a review
Renew Sust Energ Rev
(2009) - et al.
Artificial intelligence in short term electric load forecasting: a state-of-the-art survey for the researcher
Energy Convers Manag
(2003) - et al.
Overview of HVAC system simulation
Autom Constr
(2010) - et al.
A survey of energy efficient strategies for effective air conditioning
Energy Convers Manag
(2004) Theory and applications of adaptive control – a survey
Autom
(1983)- et al.
PI-control of dual duct systems: manual tuning and control loop interaction
Energy Convers Manag
(1998) - et al.
Development of an adaptive Smith predictor-based self-tuning PI controller for an HVAC system in a test room
Energy Build
(2008) - et al.
A new adaptive PI controller and its application in HVAC systems
Energy Convers Manag
(2007) - et al.
Energy optimal control of a residential space-conditioning system based on sensible heat transfer modeling
Build Environ
(2004)