Paper Conference

2020 Building Performance Analysis Conference and SimBuild co-organized by ASHRAE and IBPSA-USA


Lifting the Garage Door on Spawn, an Open-Source BEM-Controls Engine

Michael Wetter, Kyle Benne, Antoine Gautier, Thierry S. Nouidui, Agnes Ramle, Amir Roth, Hubertus Tummescheit, Stuart Mentzer, Christian Winther
Lawrence Berkeley National Laboratory, Berkeley, CA, USA
National Renewable Energy Laboratory, Golden, CO, USA
The United African University of Tanzania, Dar Es Salaam, Tanzania
Modelon AB, Lund, Sweden
US Department of Energy, Washington, DC, USA
Modelon AB, Hartford, CT, USA
Objexx Engineering, Boston, MA, USA

Abstract: Spawn is the latest whole-building energy simulation engine developed by the US Department of Energy, National Labs and industry. Whereas EnergyPlus was designed as a successor to DOE-2, Spawn is not a direct successor of - nor is it intended as an imminent replacement for - EnergyPlus. Instead, Spawn reuses parts of EnergyPlus while supporting new use cases in HVAC and controls. Spawn is intended to provide several capabilities that significantly advance beyond EnergyPlus. It is intended to support the evaluation of novel HVAC and district energy systems in a more physically realistic way. Critically, it can model control in a physically realistic way, using portable specifications that can be compiled for execution on control platforms. Spawn is also intended to support co-simulation in an intrinsic way to enable integration with third-party models. This paper describes the software architecture of Spawn from model authoring to compilation and simulation. It explains how Spawn reuses the envelope and daylighting modules of EnergyPlus and couples them to HVAC and control models from the Modelica Buildings Library using the Functional Mockup Interface (FMI) standard. It presents a number of examples that: i) validate Spawn's coupled simulation approach by comparing its results to those of EnergyPlus, ii) illustrate the Spawn methodology for modeling and simulating HVAC systems, and iii) evaluate the performance of Spawn's Quantized State System (QSS) time integration algorithms.
Pages: 518 - 525