Self-adapting Building Models For Model Predictive Control

Abstract

The capacity of intermittent power sources like wind and PV is steadily increasing. The existing balance between production and consumption is seriously affected by these new sources. Flexible demand for example in buildings is one solution to come back to a stable system. Buildings provide a huge potential for flexible electricity demand. This flexibility can be addressed by using model predictive control (MPC) with optimized scheduling for the buildings’ heating, ventilation and air conditioning (HVAC) systems. Therefore, it is necessary to have detailed simulation models of each single building. To avoid the huge effort of setting up lots of different building models, two approaches for self-adapting building models are discussed in this paper. Those approaches can be differentiated by their mathematical structure. The neural network (NN) approach is called “black-box” model. In contrast to that, the physical “white-box” model is a system of differential equations with free parameters. These models are parameterized by measured data and are developed to be used in model predictive control to forecast the building’s thermal behavior. Once the thermal behavior is predictable, the optimal schedule at minimal costs for the HVAC systems can be determined with respect to thermal comfort. The investigations show, that the difference between instance leads to low electricity prices, electrical heating systems are operated at peak load to charge the building with thermal energy. Typically, the heavier the construction of a building is the more energy can be stored. Charging a building means in this case to raise the temperature of floors, walls and ambient air. The optimization process is necessary to find out when the HVAC systems have to operate to charge the building at minimal costs. The solution of this process is the optimal schedule for the HVAC components. Depending on a flexible electricity price, the MPC calculates the optimal schedule for the HVAC systems of the building with an iterative method. To provide thermal comfort inside the building, the thermal behavior of the building is predicted with a model for each optimization step. The optimized schedule is then applied to the HVAC systems of the real building. Figure 1 provides the schematic configuration of the investigated system. self-adapting model € °C the training phase and the prediction phase is decisive for the quality of the forecast. If test and electricity price, weather data electrical HVAC system building training data are similar (e.g. same season), both models deliver satisfying results.