Autoregressive Linear Thermal Model of a Residential Forced-Air Heating System with Backpropagation Parameter Estimation Algorithm

Eric Burger, Scott Moura and David E. Culler

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2017-28
May 5, 2017

http://www2.eecs.berkeley.edu/Pubs/TechRpts/2017/EECS-2017-28.pdf

Model predictive control (MPC) strategies show great potential for improving the performance and energy efficiency of building heating, ventilation, and air-conditioning (HVAC) systems. A challenge in the deployment of such predictive thermostatic control systems is the need to learn accurate models for the thermal characteristics of individual buildings. This necessitates the development of online and data-driven methods for system identification. In this paper, we propose an autoregressive with exogenous terms (ARX) model of a building. To learn the model, we present a backpropagation approach for recursively estimating the parameters. Finally, we fit the linear model to data collected from a residential building with a forced-air heating and ventilation system and validate the accuracy of the trained model.

Advisor: David E. Culler


BibTeX citation:

@mastersthesis{Burger:EECS-2017-28,
    Author = {Burger, Eric and Moura, Scott and Culler, David E.},
    Title = {Autoregressive Linear Thermal Model of a Residential Forced-Air Heating System with Backpropagation Parameter Estimation Algorithm},
    School = {EECS Department, University of California, Berkeley},
    Year = {2017},
    Month = {May},
    URL = {http://www2.eecs.berkeley.edu/Pubs/TechRpts/2017/EECS-2017-28.html},
    Number = {UCB/EECS-2017-28},
    Abstract = {Model predictive control (MPC) strategies show great potential for improving the performance and energy efficiency of building heating, ventilation, and air-conditioning (HVAC) systems. A challenge in the deployment of such predictive thermostatic control systems is the need to learn accurate models for the thermal characteristics of individual buildings. This necessitates the development of online and data-driven methods for system identification. In this paper, we propose an autoregressive with exogenous terms (ARX) model of a building. To learn the model, we present a backpropagation approach for recursively estimating the parameters. Finally, we fit the linear model to data collected from a residential building with a forced-air heating and ventilation system and validate the accuracy of the trained model.}
}

EndNote citation:

%0 Thesis
%A Burger, Eric
%A Moura, Scott
%A Culler, David E.
%T Autoregressive Linear Thermal Model of a Residential Forced-Air Heating System with Backpropagation Parameter Estimation Algorithm
%I EECS Department, University of California, Berkeley
%D 2017
%8 May 5
%@ UCB/EECS-2017-28
%U http://www2.eecs.berkeley.edu/Pubs/TechRpts/2017/EECS-2017-28.html
%F Burger:EECS-2017-28