Grant: Discovery Grants Program - Individual
Concordia University
$24,000.00
Model-based control of high performance building systems
FY 2020-21
Summary
Model-based predictive control (MPC) for buildings has become an active research area over the last few years. MPC employs mathematical models, along with weather and occupancy forecasts, to select optimal control actions targeting an objective function (e.g., cost or load) within constraints of comfort and operational limits. Given that the term MPC tends to be associated with formal optimization algorithms, the more general expression "model-based control" (MBC) will be used herein. Appropriate models can lead to better operation strategies, and in general a coherent plan linking the design and operation of building systems. For instance, a suitable model along with “typical day” scenarios can assist in the creation of appropriate control sequences.
Developing appropriate models for MBC poses specific challenges. Existing building simulation software tools provide highly detailed models that enable interesting prospects for design applications (e.g., the comparison of millions of options). However, the requirements of a control-oriented model differ from those of a simulation model (e.g., the right time scale is different).
The project targets two closely related aspects: appropriate modelling techniques and practical model-based control solutions. As described below,
A methodology for the creation of appropriate control-oriented models for MBC applications. Desirable features of control-oriented models include adequate level of resolution; limited number of input variables; ease of calibration with recorded data; and the ability to handle stochastic representations of inputs and outputs.
Creation of a library of MBC strategies for particular cases (such as small commercial and institutional buildings, perimeter offices, and residential buildings) based on typical scenarios. These strategies, based on systematic optimization exercises, should target ease of implementation in energy management systems and smart thermostats. Other aspects of indoor environment quality (lighting, humidity control, etc.) will be considered.
In view of the interdisciplinary nature of the problem, a judicious range of techniques will be used: finite-difference modelling, using both distributed and lumped parameters; time and frequency-domain modelling; system identification techniques; data clustering of “typical days”; artificial intelligence techniques for the modelling of occupancy, human behaviour and user loads; statistical inference methods to suggest initial model parameters based on databases; and data mining and machine learning for the interpretation of data from sensors and power meters.
The 5-year project will involve the work of two PhD students, and will consist primarily of numerical simulations.
Program
Discovery Grants Program - Individual
Recipient
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
RGPIN-2016-04986
Name nom
Candanedo Ibarra, José
Department département
Buildings Group
Organization id
70
Institution établissement
Concordia University
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2020
Competition year année de concours
2016
Award amount
24000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
1512
Committee name en
Mechanical Engineering
Committee name fr
Génie mécanique
Area of application code
306
Area of application group en
Energy resources
Area of application group fr
Ressources d'énergie
Area of application en
Energy efficiency
Area of application fr
Efficacité énergétique
Research subject code
2109
Research subject group en
Mechanical engineering
Research subject group fr
Génie mécanique
Research subject en
Mechanical systems and instrumentation
Research subject fr
Instrumentation et systèmes mécaniques
Installment
inst:5-5
Partie
1
Nb partie
5
Application title
Model-based control of high performance building systems
Keyword
NSERC
Application summary
Model-based predictive control (MPC) for buildings has become an active research area over the last few years. MPC employs mathematical models, along with weather and occupancy forecasts, to select optimal control actions targeting an objective function (e.g., cost or load) within constraints of comfort and operational limits. Given that the term MPC tends to be associated with formal optimization algorithms, the more general expression "model-based control" (MBC) will be used herein. Appropriate models can lead to better operation strategies, and in general a coherent plan linking the design and operation of building systems. For instance, a suitable model along with “typical day” scenarios can assist in the creation of appropriate control sequences.
Developing appropriate models for MBC poses specific challenges. Existing building simulation software tools provide highly detailed models that enable interesting prospects for design applications (e.g., the comparison of millions of options). However, the requirements of a control-oriented model differ from those of a simulation model (e.g., the right time scale is different).
The project targets two closely related aspects: appropriate modelling techniques and practical model-based control solutions. As described below,
A methodology for the creation of appropriate control-oriented models for MBC applications. Desirable features of control-oriented models include adequate level of resolution; limited number of input variables; ease of calibration with recorded data; and the ability to handle stochastic representations of inputs and outputs.
Creation of a library of MBC strategies for particular cases (such as small commercial and institutional buildings, perimeter offices, and residential buildings) based on typical scenarios. These strategies, based on systematic optimization exercises, should target ease of implementation in energy management systems and smart thermostats. Other aspects of indoor environment quality (lighting, humidity control, etc.) will be considered.
In view of the interdisciplinary nature of the problem, a judicious range of techniques will be used: finite-difference modelling, using both distributed and lumped parameters; time and frequency-domain modelling; system identification techniques; data clustering of “typical days”; artificial intelligence techniques for the modelling of occupancy, human behaviour and user loads; statistical inference methods to suggest initial model parameters based on databases; and data mining and machine learning for the interpretation of data from sensors and power meters.
The 5-year project will involve the work of two PhD students, and will consist primarily of numerical simulations.