Physics-informed linear regression is competitive with two Machine Learning methods in residential building MPC. (15th March 2022)

• Record Type: Journal Article
• Authors: Bünning, Felix; Huber, Benjamin; Schalbetter, Adrian; Aboudonia, Ahmed; Hudoba de Badyn, Mathias; Heer, Philipp; Smith, Roy S.; Lygeros, John
• Is Part Of: Applied energy. Volume 310(2022)
• Publication Date: 2022-03-15
• Subjects: Building energy management -- Data predictive control -- Model predictive control -- Physics-informed Machine Learning -- Validation in experiment; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2021.118491
• Languages: English
• ISSNs: 0306-2619

Abstract

Abstract: Because physics-based building models are difficult to obtain as each building is individual, there is an increasing interest in generating models suitable for building MPC directly from measurement data. Machine learning methods have been widely applied to this problem and validated mostly in simulation; there are, however, few studies on a direct comparison of different models or validation in real buildings to be found in the literature. Methods that are indeed validated in application often lead to computationally complex non-convex optimization problems. Here we compare physics-informed Autoregressive–Moving-Average with Exogenous Inputs (ARMAX) models to Machine Learning models based on Random Forests and Input Convex Neural Networks and the resulting convex MPC schemes in experiments on a practical building application with the goal of minimizing energy consumption while maintaining occupant comfort, and in a numerical case study. The building has a water-based emission system and is located in temperate climate. We demonstrate that Predictive Control leads to savings between 26% and 49% of heating and cooling energy, compared to the building's baseline hysteresis controller. Moreover, we show that all model types lead to satisfactory control performance in terms of constraint satisfaction and energy reduction. However, we also see that the physics-informed ARMAX models have a lower computational burden, and a superior sample efficiency compared to the Machine Learning based models. Moreover, even if abundant training data is available, the ARMAX models have a significantly lower prediction error than the Machine Learning models, which indicates that the encoded physics-based prior of the former cannot independently be found by the latter. Highlights: Two Machine Learning methods for predictive building control are presented. A third method, based on physics-informed ARMAX models, is introduced. They are compared in experiments on a real building application. All methods reduce energy consumption by 26% to 49% compared to baseline controller. Physics-informed ARMAX outperforms the Machine Learning methods in various aspects.