A convolutional neural network prediction model for aviation nitrogen oxides emissions throughout all flight phases

Area of Science:

• Aviation environmental science
• Atmospheric chemistry
• Climate modeling

Background:

• Nitrogen oxides (NOx) emissions from aviation significantly impact atmospheric composition, particularly ozone formation in the upper troposphere.
• Existing NOx emission databases are incomplete, with scarce data for cruise phases due to sampling costs and uncertainties.
• A comprehensive emission database covering both ground and cruise phases is crucial for accurate aviation environmental impact assessments.

Purpose of the Study:

• To develop a novel prediction model for aviation nitrogen oxides (NOx) emission indices.
• To accurately predict NOx emissions for both ground and cruise phases across various engine types and fuels, including sustainable aviation fuels (SAFs).
• To establish a foundation for a more complete and accurate aviation NOx emission inventory.

Main Methods:

• Utilized convolutional neural network (CNN) technology to create a prediction model for NOx emission indices.
• Integrated data from the International Civil Aviation Organization's (ICAO) Engine Emission Database (EEDB) and in-situ measurements.
• Validated the model's accuracy by comparing predicted data against measured emissions.

Main Results:

• The CNN model demonstrated high prediction accuracy for NOx emissions in both ground (R² > 0.95) and cruise (R² > 0.9) phases.
• The model's performance surpasses traditional methods like the Boeing Fuel Flow Method 2 (BFFM2).
• Accurate prediction of NOx emissions from aircraft using SAFs was achieved.

Conclusions:

• The developed CNN-based model provides a robust tool for predicting aviation NOx emissions across the entire flight envelope.
• This facilitates the creation of more accurate aviation emission inventories, reducing uncertainties in climate and air quality impact assessments.
• The model supports research into the environmental effects of aviation, including the use of sustainable aviation fuels.