Abstract:Black-box models have demonstrated remarkable accuracy in forecasting building energy loads. However, they usually lack interpretability and do not incorporate domain knowledge, making it difficult for users to trust their predictions in practical applications. One important and interesting question remains unanswered: is it possible to use intrinsically interpretable models to achieve accuracy comparable to that of black-box models? With an aim of answering this question, this study proposes an intrinsically interpretable machine learning-based method to forecast building energy loads. It creatively combines two intrinsically interpretable machine learning algorithms: clustering decision trees and adaptive multiple linear regression. Clustering decision trees aim to automatically identify various building operation conditions, allowing for the training of multiple models tailored to each condition. It can reduce the complexity of model training data, leading to higher accuracy. Adaptive multiple linear regression is an improved regression algorithm tailored to building energy load prediction. It can adaptively modify regression coefficients according to building operations, enhancing the non-linear fitting capability of multiple linear regression. The proposed method is evaluated utilizing the operational data from an office building. The results indicate that the proposed method exhibits comparable accuracy to both random forests and extreme gradient boosting. Furthermore, it shows significantly superior accuracy, with an average improvement of 10.2%, compared with some popular black-box algorithms such as artificial neural networks, support vector regression, and classification and regression trees. As for model interpretability, the proposed method reveals that historical cooling loads are the most crucial for predicting building cooling loads under most conditions. Additionally, outdoor air temperature has a significant contribution to building cooling load prediction during the daytime on weekdays in summer and transition seasons. In the future, it will be valuable to explore integrating the laws of physics into the proposed method to further enhance its interpretability.

Limitations of machine learning for building energy prediction: ASHRAE Great Energy Predictor III Kaggle competition error analysis

The ASHRAE Great Energy Predictor III competition: Overview and results

A Machine Learning Approach for Generating and Evaluating Forecasts on the Environmental Impact of the Buildings Sector

Gradient boosting machines and careful pre-processing work best: ASHRAE Great Energy Predictor III lessons learned

Accuracy analyses and model comparison of machine learning adopted in building energy consumption prediction

Machine Learning Algorithms for Predicting Energy Consumption in Educational Buildings

Urban building energy prediction at neighborhood scale

Intrinsically Interpretable Machine Learning-Based Building Energy Load Prediction Method with High Accuracy and Strong Interpretability

Applying machine learning algorithms to group the energy performance of existing buildings by different granularity level of energy data

Using Google Trends as a proxy for occupant behavior to predict building energy consumption

Application of Machine Learning for Predicting Building Energy Use at Different Temporal and Spatial Resolution under Climate Change in USA

Automated Machine Learning-Based Building Energy Load Prediction Method

Electric energy consumption predictions for residential buildings: Impact of data-driven model and temporal resolution on prediction accuracy

Predictive Model of Energy Consumption Using Machine Learning: A Case Study of Residential Buildings in South Africa

Evaluation of Deep Learning-based prediction models in Microgrids

Optimizing Building Short-Term Load Forecasting: A Comparative Analysis of Machine Learning Models

Electrical Energy Prediction in Residential Buildings for Short-Term Horizons Using Hybrid Deep Learning Strategy

Machine Learning Methods to Increase the Energy Efficiency of Buildings

Building energy consumption prediction for residential buildings using deep learning and other machine learning techniques

Machine Learning-Based Model for Prediction of Power Consumption in Smart Grid

Developing a machine learning based building energy consumption prediction approach using limited data: Boruta feature selection and empirical mode decomposition