EFFECTS OF TIME DELAYS AND DATA INTERPOLATION METHODS ON LONGITUDINAL AERODYNAMIC DERIVATIVES ESTIMATES

Aircraft parameter estimation from flight data has reached maturity as standard procedure for experimental modeling in the aircraft industries and related research institutes, as can be seen from the large number of recent books on the subject. One of the central issues here is the existence of time delays and different data sampling rates in the flight data measurements that can, if neglected, cause accuracy loss: the aerodynamic parameter estimates will change in order to explain the delays. This report is concerned with the effect of a correction algorithm used to eliminate the time delay inherent to the Pulse Code Modulation data acquisition system, and the performance of two data interpolation scheme used to tackle the differences in sampling rates between measured data. More precisely, first part of this report analyses the influence of delays introduced by the data acquisition system, and the second part consists of a comparison between two different interpolation methods: a linear interpolation and a spline interpolation, used to deal with sub-sampled flight data, using, in both cases, the standard deviation of the estimated parameters as a comparison standard. The analysis was carried out using parameter identification by output-error with a Levenberg- Marquardt method, with corrected standard deviation for the hypothesis of colored noise. The data used herein are from real flight of a regional type aircraft.