Implementation of a Prediction Model in a Smart System for Enhancing Comfort in Dwellings

Abstract

This article introduces a novel approach to ensuring optimal comfort in residential environments, using a smart system powered by predictive modeling. At its core lies a complex algorithm, presented alongside a detailed block diagram, guiding the system’s operations, which are tailored for residential comfort. The primary focus is on the time series analysis of forecasting relative humidity—a critical parameter influencing comfort in living spaces. Among the various prediction models analyzed, a model based on the Fourier equation emerged as the most efficient, accounting for approximately 81% of variances in data. Upon validation, the model showcases an impressive relative error of just ±0.1%. The research underscores the potential of leveraging advanced forecasting in optimizing devices like dehumidifiers or air humidifiers, ensuring the desired comfort while minimizing energy consumption. This innovative integration paves the way for a smarter, more sustainable residential living experience.