A Gaussian Process Regression Approach to Model Aircra Engine Fuel Flow Rate

Œe problem of building statistical models of cyber-physical systems using operational data is addressed in this paper, using the case study of aircra‰ engines. Œese models serve as a complement to physics-based models, which may not accurately reƒect the operational performance of systems. Œe accurate modeling of fuel ƒow rate is an essential aspect of analyzing aircra‰ engine performance. In this paper, operational data from Flight Data Recorders are used to model the fuel ƒow rate. Œe independent variables are restricted to those which are obtainable from trajectory data. Treating the engine as a statistical system, an algorithm based on Gaussian Process Regression (GPR) is developed to estimate the fuel ƒow rate during the airborne phases of ƒight. Œe algorithm propagates the uncertainty in the estimates in order to determine prediction intervals. Œe proposed GPR models are evaluated for their predictive performance on an independent set of ƒights. Œe resulting estimates are also compared with those given by the Base of Aircra‰ Data (BADA) model, which is widely used in aircra‰ performance studies. Œe GPR models are shown to perform statistically signi€cantly beŠer than the BADA model. Œe GPR models also provide interval estimates for the fuel ƒow rate which reƒect the variability seen in the data, presenting a promising approach for data-driven modeling of cyber-physical systems.