Uncertainty Quantification of a Cyber-Physical System for Monitoring Aircraft Performance

This paper focuses on quantifying the uncertainties of an advanced cyber-physical system for monitoring aircraft performance, particularly fuel consumption, using a physics-based digital twin of a generic narrowbody airliner. The study highlights the limitations of current aircraft performance monitoring systems in detecting and isolating specific aerodynamic deterioration that contributes to increased fuel consumption. The cyber-physical system leverages a much larger set of sensor data and accounts for the uncertain effects of operational factors like weight variations and sensor errors. The experimental approach involves a controlled digital environment that emulates the uncertainties of real-world operating conditions. The results quantify the precision of the model, implemented as part of a digital twin, enabling a better informed and more actionable monitoring system than traditional methods. Ideally, we would like to measure changes to fuel consumption with margins of error as small as a few tenths of a percent. The paper underscores the importance of accounting for aircraft and operational effects in aerodynamics and fuel efficiency monitoring combined with big data analytics, setting a benchmark for future hybrid and data-driven models.