Sign in Subscribe
Global warming

Machine Learning Models for Biogas Potential in Sustainable Aviation: A New Study

Strategic Insights

— 1 min read

A recent research study aimed to assess the performance and emissions characteristics of different biofuel blends in aviation engines using machine learning models. The study involved testing various combinations of biofuel blends, including microalgae, biodiesel, and biogas, and evaluated their impact on energy performance and emissions reduction. The research found that the random forest model was the most effective in predicting thrust, turbine inlet temperature, and CO2 emissions, while the ridge regression model outperformed other models in predicting fuel consumption, NOx emissions, and CO emissions.

Key Takeaways:

  • The study tested four different biofuel blends: B10 (10% microalgae, 90% Jet A fuel), BB10 (10% biodiesel, 10% biogas, 80% Jet A fuel), B30 (30% microalgae, 70% Jet A fuel), and BB30 (30% biodiesel, 10% biogas, 60% Jet A fuel).
  • The machine learning models used in the study were XGBoost, random forest, and ridge regression, which were trained using actual data from the experimental study.
  • The random forest model emerged as the most effective in predicting thrust, turbine inlet temperature, and CO2 emissions, with low error and high R-squared values.
  • The ridge regression model outperformed other models in predicting fuel consumption, NOx emissions, and CO emissions.
  • The study found that the models were able to capture the lower emissions of NOx, CO, and CO2 for biodiesel blends compared to Jet A fuel.
  • The research highlighted the potential of machine learning models in supporting decision-making processes in fuel selection and engine performance optimization.

Statistics:

  • The study tested four different biofuel blends: B10, BB10, B30, and BB30.
  • 10% microalgae was used in the B10 blend, whereas 30% microalgae was used in the B30 blend.
  • The study used machine learning models to predict thrust, turbine inlet temperature, and CO2 emissions, with the random forest model outperforming other models in these areas.
  • 10% biodiesel was used in the BB10 blend, whereas 30% biodiesel was used in the BB30 blend.
  • The study found that the ridge regression model outperformed other models in predicting fuel consumption, NOx emissions, and CO emissions.

Sources:

  • Machine Learning Models for Biogas Potential In Sustainable Aviation: Xgboost, Random Forest and Ridge Regression. Aircraft Engineering and Aerospace Technology, 2025.
  • NewsRx. Studies in the Area of Biofuel Reported from GMR Institute of Technology (Machine Learning Models for Biogas Potential In Sustainable Aviation: Xgboost, Random Forest and Ridge Regression). Global Warming Focus. May 26, 2025; p 406.

Read more