Hybrid modeling approach for better identification of building thermal network model and improved prediction

TL;DR

This paper introduces a hybrid modeling approach that combines gray-box thermal network models with disturbance models, significantly improving temperature prediction accuracy for building climate control in various climates.

Contribution

The study presents a novel hybrid modeling method that effectively incorporates unmeasured disturbances into thermal network models, enhancing prediction accuracy over traditional methods.

Findings

Reduced RMSE by approximately 0.2-0.9°C in mild climates

Reduced RMSE by approximately 0.3-2°C in cold climates

Superior prediction performance on experimental laboratory data

Abstract

The gray-box modeling approach, which uses a semi-physical thermal network model, has been widely used in building prediction applications, such as model predictive control (MPC). However, unmeasured disturbances, such as occupants, lighting, and in/exfiltration loads, make it challenging to apply this approach to practical buildings. In this study, we propose a hybrid modeling approach that integrates the gray-box model with a model for unmeasured disturbance. After reviewing several system identification approaches, we systematically designed the unmeasured disturbance model with a model selection process based on statistical tests to make it robust. We generated data based on the building model calibrated by real operational data and then trained the hybrid model for two different weather conditions. The Hybrid model approach demonstrates the reduction of RMSE approximately 0.2-0.9C and 0.3-2C on 1-day ahead temperature prediction compared to the Conventional approach for mild (Berkeley, CA) and cold (Chicago, IL) climates, respectively. In addition, this approach was applied for experimental data obtained from the laboratory building to be used for the MPC application, showing superior prediction performances.