Indoor thermal comfort management: A Bayesian machine-learning approach to data denoising and dynamics prediction of HVAC systems

TL;DR

This paper presents a Bayesian machine learning framework with a deep Kalman filter for accurate indoor temperature prediction and HVAC system management, effective even with noisy data and low-quality sensors.

Contribution

It introduces a novel data-driven approach combining Bayesian and deep learning architectures, including a deep Kalman filter, for modeling indoor climate dynamics with noisy inputs.

Findings

Achieves state-of-the-art 150-minute temperature prediction with RMSE of 0.2455

Model maintains high performance with noisy data and low-accuracy sensors

Framework can be extended to demand response and equipment failure detection

Abstract

The optimal management of a building's microclimate to satisfy the occupants' needs and objectives in terms of comfort, energy efficiency, and costs is particularly challenging. This complexity arises from the non-linear, time-dependent interactions among all the variables of the control problem and the changing internal and external constraints. Focusing on the accurate modeling of the indoor temperature, we propose a data-driven approach to address this challenge. We account for thermal inertia, non-linear effects, small perturbations of the indoor climate dynamics caused by ventilation and weather variations, as well as for the stochastic nature of the control system due to the observed noise in the input signal. Since the prohibitive cost of quality data acquisition and processing limits the implementation of data-driven approaches for real-life problems, we applied a method that merges several Bayesian machine learning and deep learning architectures that are suitable for predicting complex system dynamics, while relaxing the dataset quality requirements. Our framework includes a built-in deep Kalman filter, which makes it deployable even with low-accuracy temperature sensors. It achieves state-of-the-art performance, best performing with a 150-minute prediction horizon with an RMSE of 0.2455, an MAE of 0.162, and an $R^2$ of 0.926. The model's performance remains consistent even when exposed to highly noisy data. Finally, we show how our approach can be extended to other applications including demand response event duration prediction and equipment failure detection.