Model implementation and verification of the envelope, HVAC and controller of an office building in Modelica

Abstract

Modelica is a promising open-source modelling language for simulation and derivative-based optimisation of multi-zone buildings, including detailed pressure-driven flow networks and building controllers. Such Modelica models have however not yet been demonstrated for buildings with many zones and complex HVAC systems. This paper presents and provides implementation details of the simulation model of a 32-zones building model in Modelica, with pressure-driven flow networks and a building controller, demonstrating the potential of Modelica. The modelling approach and computational aspects are presented. Furthermore, the model is calibrated and verified using on-site measurement data, which shows that, for the presented case study, HVAC can be modelled with errors up to only a few percent, while larger errors are observed for the building envelope model. A detailed building description is provided as supplementary material.

TACO, an automated toolchain for model predictive control of building systems: implementation and verification

“…A companion paper applies the TACO framework to a multi-zone office building model (Jorissen, Boydens, and Helsen 2018a), which demonstrates the performance of the framework and the energy savings potential of optimal control of advanced HVAC systems. Furthermore, it is shown that TACO efficiently scales to large problem sizes, and to complex non-linear systems (Jorissen, Boydens, and Helsen 2018b).…”

Proceedings of 14th Modelica Conference 2021, Linköping, Sweden, September 20-24, 2021

“…The most important non-linearities are the relation between flow rate and pressure in fluid flow networks, and the strongly non-linear relation between fan/pump power and flow rate. For more details about these models see (Filip Jorissen, Reynders, et al 2018;F. Jorissen, Boydens, and L. Helsen 2019;Filip Jorissen, Michael Wetter, and Lieve Helsen 2018;F.…”

Integrated Modelica Model and Model Predictive Control of a Terraced House Using IDEAS

“…The goal of TACO is to significantly reduce the engineering overhead required for developing MPCs for building applications by leveraging object oriented Modelica models. Jorissen et al (2018a) describe the detailed Modelica model of an office building and a comparison with measured data. An MPC has been developed for this model by Jorissen (2018).…”

“…Furthermore, we propose to use explicit time integrators such as Euler integration since the computation time of Euler integration scales better with the number of state variables than for implicit solvers. These tips are applied by Jorissen et al (2018a) on the model of a 32-zones office building.…”