Flight-Test Validation of a Takeoff Performance Uncertainty Model

Abstract

A Monte Carlo model designed for fixed-wing aircraft takeoff performance uncertainty quantification is benchmarked. The uses of an efficient takeoff simulator of this type range from rapid design variable and constraint sensitivity studies and large-scale conceptual level analyses to operational performance planning and real-time anomaly detection. The accuracy of the model is assessed against high-resolution flight-test data obtained through a campaign consisting of eight takeoffs flown with a specially instrumented commuter category transport aircraft: a BAe Jetstream Series 3100 twin turboprop. On all but one of the takeoffs, a close agreement is seen in terms of the takeoff distance, as predicted vs as observed, at the point of passing a 35 ft screen height; for the outlier, evidence of a sudden change in wind speed is presented as the probable cause of the discrepancy. Such studies are subject to many other sources of error and uncertainty, which are inherent in both flight-test data analysis and simulation, stemming from the highly dynamic and complex nature of this phase of the flight. The analysis presented also proposes to be a template for dealing with these issues, in a way that is applicable to other benchmarking studies.