Ralph Evins
ABSTRACT
We present a future vision for a control approach in which simple predictive models can be used to improve the performance of complex building energy systems. The philosophy is that incremental improvements upon current control systems are possible by adding small degrees of predictive capabilities at critical points. The approach follows a process of: a) gathering data on current performance; b) analysis of this data to identify key interventions; c) fitting a simplified model to enact that intervention; and d) implementing the model in the system. Rather than attempting to implement a model predictive control (MPC) scheme, a form of optimal control, a simple but predictive black-box model is used. The overall system is designed to function seamlessly in aplug-and-play fashion in conjunction with an existing building management system (BMS), which is achieved via a BACnet interface using the VOLTTRON platform. A discussion of the approach is presented along with a simple test case.
- Martín Abadi, Ashish Agarwal, Paul Barham, Eugene Brevdo, Zhifeng Chen, Craig Citro, Greg S Corrado, Andy Davis, Jeffrey Dean, Matthieu Devin, et al. 2016. Tensorflow: Large-scale machine learning on heterogeneous distributed systems. arXiv preprint arXiv:1603.04467 (2016).Google Scholar
- Venkatesh Chinde, Adam Kohl, Zhanhong Jiang, Atul Kelkar, and Soumik Sarkar. 2016. A VOLTTRON based implementation of Supervisory Control using Generalized Gossip for Building Energy Systems. (2016).Google Scholar
- Jereme Haack, Bora Akyol, Craig Allwardt, Srinivas Katipamula, Zachary Beech, Robert Lutes, Joseph Chapman, and Kyle Monson. 2016. VOLTTRONâĎć: Using distributed control and sensing to integrate buildings and the grid. In Internet of Things (WF-IoT), 2016 IEEE 3rd World Forum on. IEEE, 228--232.Google ScholarCross Ref
- Grant Hernandez, Orlando Arias, Daniel Buentello, and Yier Jin. 2014. Smart nest thermostat: A smart spy in your home. Black Hat USA (2014).Google Scholar
- Michael I Jordan and Tom M Mitchell. 2015. Machine learning: Trends, perspectives, and prospects. Science 349, 6245 (2015), 255--260.Google Scholar
- Srinivas Katipamula, Jereme Haack, George Hernandez, Bora Akyol, and Joseph Hagerman. 2016. VOLTTRON: An Open-Source Software Platform of the Future. IEEE Electrification Magazine 4, 4 (2016), 15--22.Google ScholarCross Ref
- Peter Rockett and Elizabeth Abigail Hathway. 2016. Model-predictive control for non-domestic buildings: a critical review and prospects. Building Research & Information (2016), 1--16.Google Scholar
- Kurt W Roth, Detlef Westphalen, Michael Y Feng, Patricia Llana, and Louis Quartararo. 2005. Energy impact of commercial building controls and performance diagnostics: market characterization, energy impact of building faults and energy savings potential. Prepared by TAIX LLC for the US Department of Energy. November. 412pp (Table 2-1) (2005).Google Scholar
- U.S. Department of Energy. 2008. Energy Efficiency Trends in Residential and Commercial Buildings. (2008). https://www1.eere.energy.gov/buildings/publications/pdfs/corporate/bt_stateindustry.pdfGoogle Scholar
- U.S. Energy Information Administration. 2012. Commercial Buildings Energy Consumption Survey. (2012). https://www.eia.gov/consumption/commercial/data/2012/Google Scholar
- Rayoung Yang and Mark WN ewman. 2013. Learning from a learning thermostat: lessons for intelligent systems for the home. In Proceedings of the 2013 ACM international joint conference on Pervasive and ubiquitous computing. ACM, 93--102. Google ScholarDigital Library
Recommendations
Preliminary results of model predictive control of shading systems (WIP)
SimAUD '13: Proceedings of the Symposium on Simulation for Architecture & Urban DesignShades in buildings are widely installed and are an effective technique for managing solar gains and occupant comfort. A model of a typical office space located in Ottawa, Ontario has been created and the model was developed for analysis under variable ...
Comments