%0 Conference Paper %B 1st American Modelica Conference %D 2018 %T Control Description Language %A Michael Wetter %A Milica Grahovac %A Jianjun Hu %K buildings %K controls %K hvac %X

Properly designed and implemented building control sequences can significantly reduce energy consumption. However, there is currently no process with supporting tools that allows the assessment of the performance of different control sequences, export the control sequences in a vendor-neutral format for cost estimation and for implementation on a building automation system through machine-to-machine translation, and reuse the sequences for verification during commissioning.

This paper describes a Control Description Language (CDL) that we developed to create such a process. For CDL, we selected a subset of Modelica that allows a convenient representation of control sequences, simulation of the control sequence coupled to a building energy model, and development of translators from CDL to building automation systems. To aid in the development of such translators, we created a translator from CDL to a JSON intermediate format. In future work, we seek to work with building control providers to develop translators from CDL to commercial building automation systems.

Through a case study, we show that CDL suffices for simulation-based performance assessment of two ASHRAE-published control sequences for a variable air volume flow system of an office building. Moreover, the case study showed that merely due to differences in the control sequences, annual HVAC energy use was reduced by 30%. This difference is larger than the accuracy required when comparing different HVAC systems, thereby questioning the current practice of idealizing control sequences in building energy simulations, and demonstrating the importance of ensuring that the control sequence used during design simulations corresponds to the control sequence that will be implemented in the real building

%B 1st American Modelica Conference %8 08/2018 %G eng %U https://simulationresearch.lbl.gov/wetter/download/2018-americanModelica-WetterGrahovacHu.pdf %2 LBNL-2001219 %0 Conference Paper %B 2018 Building Performance Modeling Conference and SimBuild co-organized by ASHRAE and IBPSA-USA %D 2018 %T OpenBuildingControl: Modeling Feedback Control as a Step Towards Formal Design, Specification, Deployment and Verification of Building Control Sequences %A Michael Wetter %A Jianjun Hu %A Milica Grahovac %A Brent Eubanks %A Philip Haves %X

This paper presents ongoing work to develop tools and a process that will allow building designers to instantiate control sequences, configure them for their project, assess their performance in closed loop building energy simulation, and then export these sequences (i) for the control provider to bid on the project and to implement the sequences through machine-to-machine translation, and (ii) for the commissioning agent to verify their correct implementation.

The paper reports on the following: (i) The specification of a Control Description Language, (ii) its use to implement a subset of the ASHRAE Guideline 36 sequences, released as part of the Modelica Buildings library, (iii) its use in annual closed-loop simulations of a variable air- volume flow system, and (iv) lessons learned regarding simulation of closed-loop control.

In our case study, the Guideline 36 sequences yield 30% lower annual site HYAC energy use, under comparable comfort, than sequences published earlier by ASHRAE. The 30% differences in annual HYAC energy consumption due to changes in the control sequences raises the question of whether the idealization of control sequences that is common practice in today's building energy simulation leads to trustworthy energy use predictions.

%B 2018 Building Performance Modeling Conference and SimBuild co-organized by ASHRAE and IBPSA-USA %C Chicago, IL %8 09/2018 %G eng %U https://www.ashrae.org/File%20Library/Conferences/Specialty%20Conferences/2018%20Building%20Performance%20Analysis%20Conference%20and%20SimBuild/Papers/C107.pdf