Detailed White-Box Non-Linear Model Predictive Control for Scalable Building HVAC Control

Abstract

Grey-box and black-box MPC approaches for building HVAC applications often use lumped, low-order models with a low level of detail. While such models require smaller computation times, their accuracy is limited and there are practical constraints related to data collection, how to deal with multi-zone buildings and they often do not explicitly model the building HVAC equipment. In this paper we present an alternative approach based on detailed white-box models. TACO, a custom toolchain that builds upon physics-based Modelica models and JModelica, is used to efficiently solve the resulting optimisation problems. This paper presents a case study model of 79 zones and OCP results for this case study are discussed, demonstrating the high potential of detailed white-box MPC.