Weather and load retrieval improves short-term building energy forecasts
A Model-Agnostic Correction Framework for Short-Term Building Load Forecasting
Information RetrievalMachine Learning
Summary
Predicting how much energy a building will use soon is tricky because weather and how the building is used can change suddenly. The authors created a method called WeL-RAC that looks at past similar weather and building energy patterns to fix and improve these predictions right when they are needed. This method works with many existing forecasting tools without changing them and helps make energy use predictions more accurate and understandable. They tested WeL-RAC on many buildings and forecast times and found it often made predictions better.
What this means in practice
- •For building energy managers: Improve the accuracy of short-term energy use forecasts by correcting predictions using historical data matched to current weather and load conditions.
- •For renewable energy operators: Use corrected building load forecasts to better balance supply and demand with variable renewable energy sources like solar and wind.
Authors
Shiqiao Zhou, Huanbo Lyu, Holger Schöner, Zipeng Wu, Jessica Deuschel, Freya Wentzel, IAG Wilson, Shuo Wang
Abstract
Short-term building load forecasting remains challenging because load patterns can change rapidly with weather and operating conditions. Data-driven forecasting models may become less reliable when the current operating regime is rare or poorly represented during training, even when similar situations exist elsewhere in the historical record. Recent advances in retrieval-augmented time series forecasting offer a promising way to adapt predictions to changing conditions by directly reusing relevant historical cases at inference time. Motivated by this idea, we propose WeL-RAC, a model-agnostic Weather-Load Retrieval-Augmented Correction framework that explicitly retrieves such evidence at prediction time. WeL-RAC retrieves historical cases that are similar to the current condition based on both recent load states and horizon-aligned weather forecasts. Their observed future load trajectories are then aggregated and used to correct the initial prediction, without retraining or modifying the base forecaster. We evaluate WeL-RAC on 31 building cases across three datasets, eight forecasting backbones, and four horizons from 1 to 18 hours. The framework improves the corresponding base forecasters in 90 of 96 dataset-level comparisons. Beyond accuracy, success and failure case studies further illustrate the framework's interpretability by tracing how retrieved historical evidence improves or degrades forecasts. These findings support explicit historical retrieval as an effective and interpretable complement to the building load forecasting.