Exploring the thermal dynamics of residential spaces using sparse regression models and their application in energy-efficient control

Abstract

In this study, we present a methodological framework for the thermodynamic modeling of a residential space and propose an automation and control strategy built upon this description. In addition, based on the collected measurement data, we discuss the results obtained at each stage of the implementation process. The core of our research is a specialized, algorithm‑based dynamic system identification method used to model the complex thermodynamic behavior of residential environments, with particular emphasis on the thermal inertia of walls and interior furnishings. In response to rising energy costs, we complement the physical constraints of the system with a financial constraint, and these considerations together form the basis of our proposed optimization strategy.