Chun Fu

Clayton Miller, Liu Hao, Chun Fu

The ASHRAE Great Energy Predictor III (GEPIII) competition was held in late 2019 as one of the largest machine learning competitions ever held focused on building performance. It was hosted on the Kaggle platform and resulted in 39,402 prediction submissions, with the top five teams splitting $25,000 in prize money. This paper outlines lessons learned from participants, mainly from teams who scored in the top 5% of the competition. Various insights were gained from their experience through an online survey, analysis of publicly shared submissions and notebooks, and the documentation of the winning teams. The top-performing solutions mostly used ensembles of Gradient Boosting Machine (GBM) tree-based models, with the LightGBM package being the most popular. The survey participants indicated that the preprocessing and feature extraction phases were the most important aspects of creating the best modeling approach. All the survey respondents used Python as their primary modeling tool, and it was common to use Jupyter-style Notebooks as development environments. These conclusions are essential to help steer the research and practical implementation of building energy meter prediction in the future.

Chun Fu, Clayton Miller

In recent years, the availability of larger amounts of energy data and advanced machine learning algorithms has created a surge in building energy prediction research. However, one of the variables in energy prediction models, occupant behavior, is crucial for prediction performance but hard-to-measure or time-consuming to collect from each building. This study proposes an approach that utilizes the search volume of topics (e.g., education} or Microsoft Excel) on the Google Trends platform as a proxy of occupant behavior and use of buildings. Linear correlations were first examined to explore the relationship between energy meter data and Google Trends search terms to infer building occupancy. Prediction errors before and after the inclusion of the trends of these terms were compared and analyzed based on the ASHRAE Great Energy Predictor III (GEPIII) competition dataset. The results show that highly correlated Google Trends data can effectively reduce the overall RMSLE error for a subset of the buildings to the level of the GEPIII competition's top five winning teams' performance. In particular, the RMSLE error reduction during public holidays and days with site-specific schedules are respectively reduced by 20-30% and 2-5%. These results show the potential of using Google Trends to improve energy prediction for a portion of the building stock by automatically identifying site-specific and holiday schedules.

Clayton Miller, Bianca Picchetti, Chun Fu et al.

Research is needed to explore the limitations and potential for improvement of machine learning for building energy prediction. With this aim, the ASHRAE Great Energy Predictor III (GEPIII) Kaggle competition was launched in 2019. This effort was the largest building energy meter machine learning competition of its kind, with 4,370 participants who submitted 39,403 predictions. The test data set included two years of hourly whole building readings from 2,380 meters in 1,448 buildings at 16 locations. This paper analyzes the various sources and types of residual model error from an aggregation of the competition's top 50 solutions. This analysis reveals the limitations for machine learning using the standard model inputs of historical meter, weather, and basic building metadata. The errors are classified according to timeframe, behavior, magnitude, and incidence in single buildings or across a campus. The results show machine learning models have errors within a range of acceptability (RMSLE_scaled =< 0.1) on 79.1% of the test data. Lower magnitude (in-range) model errors (0.1 < RMSLE_scaled =< 0.3) occur in 16.1% of the test data. These errors could be remedied using innovative training data from onsite and web-based sources. Higher magnitude (out-of-range) errors (RMSLE_scaled > 0.3) occur in 4.8% of the test data and are unlikely to be accurately predicted.