Equation-based languages– A new paradigm for building energy modeling, simulation and optimization

TitleEquation-based languages– A new paradigm for building energy modeling, simulation and optimization
Publication TypeJournal Article
Year of Publication2015
AuthorsWetter, Michael, Marco Bonvini, and Thierry Stephane Nouidui
Abstract

Most of the state-of-the-art building simulation programs implement
models in imperative programming languages. This complicates
modeling and excludes the use of certain efficient methods for simulation
and optimization. In contrast, equation-based modeling languages
declare relations among variables, thereby allowing the use of
computer algebra to enable much simpler schematic modeling and to
generate efficient code for simulation and optimization.
We contrast the two approaches in this paper. We explain how
such manipulations support new use cases. In the first of two examples,
we couple models of the electrical grid, multiple buildings,
HVAC systems and controllers to test a controller that adjusts building
room temperatures and PV inverter reactive power to maintain
power quality. In the second example, we contrast the computing
time for solving an optimal control problem for a room-level model
predictive controller with and without symbolic manipulations. Exploiting
the equation-based language led to 2, 200 times faster solution

LBNL Report Number

LBNL-1003383