Model-predictive control of mixed-mode buildings with rule extraction

May-Ostendorp, Peter; Henze, Gregor P.; Corbin, Charles D.; Rajagopalan, Balaji; Felsmann, Clemens

doi:10.1016/j.buildenv.2010.08.004

Cited by 151 publications

(69 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although optimisation studies are most commonly performed in the early-design stage, where the majority of design decisions have yet to be made, optimisation approaches can be equally useful in the late-design stages for selecting and fine-tuning control strategies and HVAC design and during building operations to best select building control based on model-predictive control strategies [21][22][23][24]. The most appropriate search algorithms and modelling approaches vary depending on the application area, but the suitable application area for optimisation methodologies related to building design and control is vast and constantly evolving.…”

Section: Importance Of Bpomentioning

confidence: 99%

Assessing gaps and needs for integrating building performance optimization tools in net zero energy buildings design

Attia

Hamdy

O’Brien³

et al. 2013

Energy and Buildings

369

192

View full text Add to dashboard Cite

This paper summarizes a study undertaken to reveal potential challenges and opportunities for integrating optimisation tools in Net Zero Energy Buildings (NZEB) Design. The paper reviews current trends in simulationbased Building Performance Optimisation (BPO) and outlines major criteria for optimisation tools selection and evaluation. This is based on analyzing user's needs for tools capabilities and requirement specifications. The review is carried out by means of a literature review of 165 publications and interviews with 28 optimisation experts. The findings are based on an inter-group comparison between experts. The aim is to assess the gaps and needs for integrating BPO tools in NZEB Design. The findings indicate existing limitations including model uncertainty, computation time, difficulty of use and steep learning curve. Some future directions anticipated or needed for improvement of current tools are presented.

show abstract

Section: Importance Of Bpomentioning

confidence: 99%

Assessing gaps and needs for integrating building performance optimization tools in net zero energy buildings design

Attia

Hamdy

O’Brien³

et al. 2013

Energy and Buildings

369

192

View full text Add to dashboard Cite

show abstract

“…Spindler and Norford [13] and May-Ostendorp et al [14] investigated the capability of using an MPC in mixed-mode buildings for energy conservation. Because of the computational cost associated with general MPCs, an offline MPC was recommended followed by the logistic regression on the data to extract some simple rules for real implementation [14].…”

Section: Introductionmentioning

confidence: 99%

Multi-structural fast nonlinear model-based predictive control of a hydronic heating system

Loftness

2013

Building and Environment

View full text Add to dashboard Cite

“…Rule extraction for binary inputs in a similar context has been considered in [9] with a logistic regression model. Here, we propose to use a boosting algorithm (ADABOOST, [10]) from the field of machine learning.…”

Section: Introductionmentioning

confidence: 99%

Learning near-optimal decision rules for energy efficient building control

Domahidi

Ullmann

Morari

et al. 2012

2012 IEEE 51st IEEE Conference on Decision and Control (CDC)

View full text Add to dashboard Cite

Abstract-Recent studies suggest that advanced optimization based control methods such as model predictive control (MPC) can increase energy efficiency of buildings. However, adoption of these methods by industry is still slow, as building operators are used to working with simple controllers based on intuitive decision rules that can be tuned easily on-site. In this paper, we suggest a synthesis procedure for rule based controllers that extracts prevalent information from simulation data with MPC controllers to construct a set of human readable rules while preserving much of the control performance. The method is based on the ADABOOST algorithm from the field of machine learning. We focus on learning binary decisions, considering also the ranking and selection of measurements on which the decision rules are based. We show that this feature selection is useful for both complexity reduction and decreasing investment costs by pruning unnecessary sensors. The proposed method is evaluated in simulation for six different case studies and is shown to maintain the high performance of MPC despite the tremendous reduction in complexity.

show abstract

Model-predictive control of mixed-mode buildings with rule extraction

Cited by 151 publications

References 8 publications

Assessing gaps and needs for integrating building performance optimization tools in net zero energy buildings design

Assessing gaps and needs for integrating building performance optimization tools in net zero energy buildings design

Multi-structural fast nonlinear model-based predictive control of a hydronic heating system

Learning near-optimal decision rules for energy efficient building control

Contact Info

Product

Resources

About