Uncertainty Analysis of Test Data at the NRC 1.5-Meter Trisonic Wind Tunnel

Abstract

An uncertainty analysis algorithm has been developed at the National Research Council of Canada (NRC) 1.5 meter trisonic wind tunnel in Ottawa, Ontario, in order to assess the data quality in the half-model experiments carried out by Bombardier Aerospace. The analysis follows the methodology recommended by the AIAA Standard S-071A-1999 to propagate the systematic and random errors in the measured variables through the complete data reduction routine, in order to calculate the overall uncertainties of the tunnel freestream parameters and the key aerodynamic force and moment coefficients in the form of a 95-percent confidence interval. Detailed breakdowns of the total uncertainties have been obtained to separately investigate the systematic and random components, and to identify the variables that have the largest contribution to the uncertainties. With the implementation of the developed algorithm, the uncertainties of the results are precisely quantified at every measurement point throughout an experiment. Hence, the analysis can be performed over a range of test conditions, aircraft model configurations, and angles of attack. The data of the most recent Bombardier experiments have been used to present the outcomes. The analyses reveal that the freestream properties are achieved with high accuracy at the NRC Wind tunnel, as the uncertainties in the freestream Mach number and dynamic pressure are respectively limited to 0.4% and 0.7% of the desired nominal values. The uncertainties in the coefficients of lift and drag vary over the pitch sweep during a test due to the functional dependence on the normal and axial force, as well as the angle of attack. The magnitudes and the component breakdowns of the uncertainties in these coefficients also vary with the tunnel operating condition and from one test article to another. Nevertheless, the force and pressure measurements consistently appear to be the dominant contributors. Hence, any attempts to improve the data quality should focus on the half-model balance and the pressure measurement instruments to reduce their uncertainty. The analysis results have been validated through an alternative error propagation methodology, and also by comparing the calculated random uncertainties to the observed variability in the result parameters. However, the algorithm can benefit from more comprehensive estimates of the bias and precision errors in the measurements of the force balance.