Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

EnsembleCI: Ensemble Learning for Carbon Intensity Forecasting

Leyi Yan, Linda Wang, Sihang Liu, Yi Ding

TL;DR

This paper tackles the challenge of forecasting carbon intensity (CI) across regional grids with diverse energy mixes, where long-horizon accuracy is crucial for load shifting and scheduling. It introduces EnsembleCI, an end-to-end ensemble framework that stacks multiple sublearners (e.g., LightGBM, CatBoost, NN) via two-stage stacking and ensemble selection to adapt weights by grid. Across 11 grids and up to 96 hours ahead, EnsembleCI consistently outperforms the state-of-the-art CarbonCast, achieving substantial reductions in prediction error and greater robustness on longer horizons, while revealing region-specific feature importance through permutation analysis. The work enhances interpretability and practical applicability for grid operators, enabling more reliable and scalable CI forecasting for sustainable energy management.

Abstract

Carbon intensity (CI) measures the average carbon emissions generated per unit of electricity, making it a crucial metric for quantifying and managing the environmental impact. Accurate CI predictions are vital for minimizing carbon footprints, yet the state-of-the-art method (CarbonCast) falls short due to its inability to address regional variability and lack of adaptability. To address these limitations, we introduce EnsembleCI, an adaptive, end-to-end ensemble learning-based approach for CI forecasting. EnsembleCI combines weighted predictions from multiple sublearners, offering enhanced flexibility and regional adaptability. In evaluations across 11 regional grids, EnsembleCI consistently surpasses CarbonCast, achieving the lowest mean absolute percentage error (MAPE) in almost all grids and improving prediction accuracy by an average of 19.58%. While performance still varies across grids due to inherent regional diversity, EnsembleCI reduces variability and exhibits greater robustness in long-term forecasting compared to CarbonCast and identifies region-specific key features, underscoring its interpretability and practical relevance. These findings position EnsembleCI as a more accurate and reliable solution for CI forecasting. EnsembleCI source code and data used in this paper are available at https://github.com/emmayly/EnsembleCI.

EnsembleCI: Ensemble Learning for Carbon Intensity Forecasting

TL;DR

This paper tackles the challenge of forecasting carbon intensity (CI) across regional grids with diverse energy mixes, where long-horizon accuracy is crucial for load shifting and scheduling. It introduces EnsembleCI, an end-to-end ensemble framework that stacks multiple sublearners (e.g., LightGBM, CatBoost, NN) via two-stage stacking and ensemble selection to adapt weights by grid. Across 11 grids and up to 96 hours ahead, EnsembleCI consistently outperforms the state-of-the-art CarbonCast, achieving substantial reductions in prediction error and greater robustness on longer horizons, while revealing region-specific feature importance through permutation analysis. The work enhances interpretability and practical applicability for grid operators, enabling more reliable and scalable CI forecasting for sustainable energy management.

Abstract

Carbon intensity (CI) measures the average carbon emissions generated per unit of electricity, making it a crucial metric for quantifying and managing the environmental impact. Accurate CI predictions are vital for minimizing carbon footprints, yet the state-of-the-art method (CarbonCast) falls short due to its inability to address regional variability and lack of adaptability. To address these limitations, we introduce EnsembleCI, an adaptive, end-to-end ensemble learning-based approach for CI forecasting. EnsembleCI combines weighted predictions from multiple sublearners, offering enhanced flexibility and regional adaptability. In evaluations across 11 regional grids, EnsembleCI consistently surpasses CarbonCast, achieving the lowest mean absolute percentage error (MAPE) in almost all grids and improving prediction accuracy by an average of 19.58%. While performance still varies across grids due to inherent regional diversity, EnsembleCI reduces variability and exhibits greater robustness in long-term forecasting compared to CarbonCast and identifies region-specific key features, underscoring its interpretability and practical relevance. These findings position EnsembleCI as a more accurate and reliable solution for CI forecasting. EnsembleCI source code and data used in this paper are available at https://github.com/emmayly/EnsembleCI.
Paper Structure (11 sections, 5 figures, 3 tables)

This paper contains 11 sections, 5 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Background and Motivation
  3. The State-of-the-Art: CarbonCast
  4. Why does CarbonCast not perform well consistently across grids?
  5. Can other methods surpass CarbonCast?
  6. EnsembleCI
  7. Evaluation
  8. Methodology
  9. Prediction Results
  10. Interpreting Features
  11. Conclusion

Figures (5)

  • Figure 1: The architecture design of CarbonCast maji2022carboncastmaji2023multi.
  • Figure 2: Averaged hourly prediction accuracy of CarbonCast for 11 regional grids over a 4-day horizon.
  • Figure 3: Energy source mixes for 11 regional grids in 2024.
  • Figure 4: Prediction accuracy of different methods. The best method for each grid on each day is marked with a star.
  • Figure 5: The architecture design of EnsembleCI.