A novel approach to forecast global CO₂ emission using Bat and Cuckoo optimization algorithms

Area of Science:

• Environmental Science
• Computational Intelligence
• Energy Economics

Background:

• Global CO2 emissions significantly impact climate change.
• Accurate forecasting of CO2 emissions is crucial for effective environmental policy.
• Energy consumption patterns are primary drivers of CO2 emissions.

Purpose of the Study:

• To apply Bat and Cuckoo optimization algorithms for forecasting global CO2 emissions.
• To develop and compare linear and exponential models for CO2 emission estimation.
• To evaluate the performance of metaheuristic algorithms in predicting emissions based on energy consumption.

Main Methods:

• Utilized Bat Algorithm (BA) and Cuckoo Optimization Algorithm (COA) for model development.
• Developed both linear and exponential models for CO2 emission forecasting.
• Trained models using historical data (1980-2013) and tested on recent data (2014-2018).
• Evaluated model performance using Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE).

Main Results:

• Both BA and COA models showed good agreement in forecasting global CO2 emissions.
• The Cuckoo Optimization Algorithm's exponential model (COA-GCO_2 exponential) outperformed other developed models.
• The metaheuristic approach provided a reliable method for estimating emissions from oil, natural gas, and coal consumption.

Conclusions:

• The study recommends the Cuckoo exponential model as a reliable tool for forecasting global CO2 emissions.
• The developed methodology offers a valuable approach for climate policy decision-making and environmental management.
• This research provides a benchmark dataset and methodology for future studies on greenhouse gas emissions analysis.