Our heavyweight helicopter equal in the world does not have
In Rostov started production of the most load-lifting rotary-wing car The Russian holding «Helicopt[...]
ANALYSIS OF LARGE AMPLITUDE MANEUVER DATA
The usual procedure for aircraft parameter estimation is to perturb it slightly from its trim position by deflecting one or more of its control surfaces and gathering flight data from such small amplitude maneuvers. However, in practice it may not be possible to trim the aircraft at certain test points in the flight envelope. Under such conditions, large amplitude maneuvers (LAMs) are more useful in determining the aerodynamic derivatives. The approach is not valid in the region where stability and control derivatives vary rapidly. The results of the LAM are generated for the fighter aircraft FA2 for the conditions: (1) simulated longitudinal LAM data at two flight conditions and analyzed using the SMLR (stepwise multiple regression method [1]) without partitioning—here the entire data set is handled as a single LAM; (2) the data as in (1) but analyzed using the model error method (MEM [1]); and (3) the real flight data (*) analyzed using the previous two approaches [28]. The results are shown in Table 9.13.
The other method consists of partitioning the LAM that covers a large AOA range into several bins, each of which spans a smaller range of AOA [29]. These data would not be contiguous and therefore the regression method is employed to
 |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
TABLE 9.13 Longitudinal LAM Data Results Trim
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
estimate derivatives. This approach was also used for the analysis of the data mentioned earlier, and the detailed results can be found in Ref. [28].
For LD data generation from LAMs (in an RTS), the modes were excited in roll and yaw axes by giving aileron and rudder inputs superimposed on a steadily increasing AOA due to slow but steady increase in elevator deflection, thereby generating LAM data. Normal practice is to estimate linear derivative models but, if necessary, the SMLR approach can be used to determine the model structure with higher-order terms for better representation of aircraft dynamics. A pilot flew an RTS (Chapter 6) and persistently excited the roll and yaw motions at increasing AOA in pullup. The data were generated for Mach numbers 0.4 and 0.6 with AOA variation between 2° and 20° [30]. Significant variation in yaw during lateral stick inputs, possibly caused by aileron-rudder interconnect, was seen in the time – history plots. Regression and data partitioning techniques were used for estimation of the LD derivatives. For each Mach, the data were partitioned into bins, where each bin corresponds to a mean value of AOA: at Mach 0.4, the total variation in AOA is from 2° to 20°. The data from 2° to 4° are put into one group; the data from 4° to 6° are put into another group, and so on until all the data are accounted for. The data in a group would have resulted from different portions of a single maneuver or from certain portions of several other maneuvers. These data in each group were analyzed using regression. Each bin should have sufficient number of data points for successful estimation of derivatives and should have sufficient variations in the aircraft motion variables so that accurate estimation of derivatives is possible. Several criteria, e. g., the number of data points in each bin, the CRBs of the estimated derivatives, and the percentage fit error, were evaluated to check the quality of the estimates. The derivatives C„b, C/, C„v and C/ were estimated and cross-plotted with the reference derivatives. The results were very encouraging. One compound derivative (for the flight condition Mach number = 0.6 and with auto slat) C = C^ + Q tan a is shown Figure 9.6 [30]. This RTS exercise established that the approach if followed would work for the data gathered from real flight tests. Using this approach it is possible to determine the aerodynamic derivatives at flight conditions that cannot be covered during 1g trim flight.
