From RNNs to Foundation Models: An Empirical Study on Commercial Building Energy Consumption

TL;DR

This study investigates how dataset heterogeneity affects the accuracy of energy consumption forecasting in commercial buildings, highlighting the importance of model architecture and the potential of foundation models.

Contribution

It provides an empirical analysis of the impact of data diversity on forecasting models and evaluates the effectiveness of fine-tuned foundation models in this context.

Findings

Dataset heterogeneity impacts forecasting accuracy more than model size.

Fine-tuned foundation models perform competitively despite higher computational costs.

Model architecture and data diversity are key factors influencing performance.

Abstract

Accurate short-term energy consumption forecasting for commercial buildings is crucial for smart grid operations. While smart meters and deep learning models enable forecasting using past data from multiple buildings, data heterogeneity from diverse buildings can reduce model performance. The impact of increasing dataset heterogeneity in time series forecasting, while keeping size and model constant, is understudied. We tackle this issue using the ComStock dataset, which provides synthetic energy consumption data for U.S. commercial buildings. Two curated subsets, identical in size and region but differing in building type diversity, are used to assess the performance of various time series forecasting models, including fine-tuned open-source foundation models (FMs). The results show that dataset heterogeneity and model architecture have a greater impact on post-training forecasting performance than the parameter count. Moreover, despite the higher computational cost, fine-tuned FMs demonstrate competitive performance compared to base models trained from scratch.