A Comparative Study of Machine Learning, Deep Learning, and Retrieval-Augmented Generation in Energy Consumption Forecasting
Abstract
Background: Accurate energy consumption forecasting is important for power system planning, load balancing, cost control, and sustainable energy management. Machine learning and deep learning methods have improved data-driven forecasting, but many models still provide limited transparency about the contextual evidence behind their predictions. Retrieval-Augmented Generation (RAG) offers a way to incorporate retrieved contextual information into forecasting workflows, but its practical value for energy consumption forecasting remains underexplored. Objectives: This thesis compares baseline, classical statistical, machine learning, deep learning, Transformer-based, and retrieval-augmented forecasting approaches for multiregion energy consumption forecasting. The objective is to evaluate forecasting accuracy, computational trade-offs, horizon-wise behavior, and interpretability within a unified experimental framework. Methods: A comparative forecasting pipeline was developed using chronological train-validation-test splitting, multiple direct forecasting horizons, engineered temporal and contextual features, and representative models from each forecasting family. Forecasting performance was evaluated using RMSE, MAE, and MAPE, with RMSE used as the primary ranking metric. Interpretability and diagnostic behavior were examined using feature-importance analysis, saliency observations, retrieval diagnostics, forecast-error analysis, and accuracy-compute trade-off evaluation.…
How strong is the evidence?
Evidence Hub reads this as a thesis, done in computer models (in silico), not in people. Its evidence level is 5 of 5 (Very low · case / narrative / animal / in vitro / in silico). Case reports, narrative reviews, and animal, cell or computer-model studies give the weakest evidence for human health.
Read the paper
Related studies
- Knowledge Management Perspective of Generative Artificial Intelligence
- Navigating Generative Artificial Intelligence Promises and Perils for Knowledge and Creative Work
- Characterizing the evolving research on enterprise content management
- Large language models and multimodal foundation models for precision oncology
- From Generative AI to Extended Reality: Multidisciplinary Perspectives on the Challenges, Opportunities, and Future of Educational Computing
- Generative Artificial Intelligence: A New Engine for Advancing Environmental Science and Engineering
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.