A tracking control method for electricity-carbon emission forecasting.

Area of Science:

• Environmental Science
• Data Science
• Climate Modeling

Background:

• Accurate carbon emission forecasting is crucial for effective climate change mitigation strategies.
• Existing prediction models often face challenges with accuracy, lag, and computational efficiency.
• Integrating diverse data sources can potentially improve predictive performance.

Purpose of the Study:

• To introduce a novel carbon emission prediction method.
• To enhance forecasting accuracy and minimize prediction lag.
• To reduce computational time compared to established models.

Main Methods:

• Developed a novel carbon emission prediction method based on tracking control.
• Incorporated historical CO2 emission prediction errors.
• Utilized feed-forward integration of electricity consumption data.
• Performed comparative analysis with Long Short-Term Memory (LSTM) and Autoregressive Distributed Lag (ARDL) models using Python.

Main Results:

• The proposed method demonstrated a substantial reduction in prediction errors compared to relying solely on electricity data.
• Significant reduction in computational time was observed when compared to LSTM models.
• The method effectively integrates historical errors and electricity consumption for improved forecasting.

Conclusions:

• The novel tracking control method offers superior accuracy and efficiency for carbon emission prediction.
• Findings provide a valuable reference for policymakers and researchers in climate change mitigation.
• The study highlights the potential of integrating prediction errors for enhanced forecasting.