Neural Network-Based Building Energy Models for Adapting to Post-Occupancy Conditions: A Case Study for Florida

Abstract

Building energy models (BEMs) are developed by subject matter experts during the design phase to help with decision making for achieving a more energy-efficient design. A BEM that is created based on “as-designed” condition to predict building energy consumption can become much less accurate during the lifetime of the building given the potential changes to the “in-operation” conditions. While BEMs can be adjusted to address operational changes, the end-user (i.e. building owners) usually do not possess the knowledge to update physics-based models (e.g., eQuest) and therefore the initial BEM may no longer be useful to them. In the present paper, an approach is proposed and assessed through which a physics-based model is developed using eQuest and simulated for several different operating conditions. The resulting data are then used for training an artificial neural network (ANN) which serves as a simple and data-driven model for prediction of building energy consumption in response to changes in operating conditions. A case study is performed for a building in Melbourne, FL to explore the changes occurred in the building schedule of operation during COVID-19 pandemic and it's impact on the performance of BEM. The trained ANN is tested against the actual measured data for energy consumption under different scenarios and good agreement between the results are found. The approach presented can be used to establish data-driven BEMs that remain accurate in response to sudden changes in building operating conditions.