Forecasting CO₂ emissions in Hebei, China, through moth-flame optimization based on the random forest and extreme...

Area of Science:

• Environmental Science
• Climate Science
• Data Science

Background:

• Rising carbon dioxide (CO2) emissions are a primary driver of global warming and climate change.
• Accurate CO2 emission trend analysis is vital for developing effective energy-saving and emission reduction strategies.
• Understanding emission patterns is essential for mitigating impacts on human development and global ecosystems.

Purpose of the Study:

• To develop and validate a novel hybrid model for enhanced carbon dioxide emission forecasting.
• To analyze influential factors affecting CO2 emissions and improve prediction accuracy.
• To provide a tool for informing climate change mitigation policies.

Main Methods:

• A hybrid model integrating Random Forest (RF) for factor analysis and Extreme Learning Machine (ELM) for prediction.
• Moth-Flame Optimization (MFO) algorithm to optimize the initial weights and biases of the ELM model.
• Empirical validation using a case study from Hebei Province, China (1995-2015).

Main Results:

• The proposed hybrid RF-ELM model, optimized with MFO, demonstrated superior performance in CO2 emission prediction compared to existing models.
• The model effectively identified key factors influencing carbon dioxide emissions.
• The study confirmed the model's potential to significantly enhance the accuracy of CO2 emission forecasting.

Conclusions:

• The hybrid RF-ELM model with MFO optimization offers a robust and accurate approach for carbon dioxide emission forecasting.
• This predictive capability can support the implementation of targeted emission reduction measures.
• The findings contribute to better climate change management and policy development.