1、. u t - “ 7 ,ANDNASA TECHNICAL NOTE NASA TN D-7455v-J(NASA-iN-D-7455) LATERAL STATIC AND N74-20666- DYNANIC AEPODYNANIC PARANETES OF THEKESTREL AIRCRAFT (XV-6A) EXTRACTED FROMrn FLIGHI DA2A (NASA) 45 p HC $3.25 UnclasCSCL 01C H1/02 36238ZZ93 “x-LATERAL STATIC AND DYNAMIC AERODYNAMICPARAMETERS OF THE
2、 KESTREL AIRCRAFT(XV-6A) EXTRACTED FROM FLIGHT DATAby William T. Suit and James L. WilliamsLangley Research CenterHampton, Va. 23665ATNA7ARNATIANDPA ADMINISTRATION WASHINGTON D. C APRIL 1974NATIONAL AERONAUTICS AND SPACE ADMINISTRATION WASHINGTON, D. C. APRIL 1974Provided by IHSNot for ResaleNo repr
3、oduction or networking permitted without license from IHS-,-,-1. Report No. 2. Government Accession No. 3. Recipients Catalog No.NASA TN D-74554. Title and Subtitle 5. Report DateLATERAL STATIC AND DYNAMIC AERODYNAMIC April 1974PARAMETERS OF THE KESTREL AIRCRAFT (XV-6A) 6. Performing Organization Co
4、deEXTRACTED FROM FLIGHT DATA7. Author(s) 8. Performing Organization Report No.L-9176William T. Suit and James L. Williams10. Work Unit No.9. Performing Organization Name and Address 501-26-05-02NASA Langley Research Center 11. Contract or Grant No.Hampton, Va. 2366513. Type of Report and Period Cove
5、red12. Sponsoring Agency Name and Address Technical NoteNational Aeronautics and Space Administration 14. Sponsoring Agency CodeWashington, D.C. 2054615. Supplementary Notes16. AbstractFlight test data have been used to extract the lateral static and dynamic aerodynamicparameters of the Kestrel airc
6、raft. The aircraft configurations included thrust-jet anglesof 00, 150, and 300, and the test Mach numbers were 0.43, 0.62, and 0.82. The resultsshowed that most of the parameters varied linearly with trim normal-force coefficient.The directional stability parameter Cn showed a small increase with i
7、ncreasing trimnormal-force coefficient and also with nozzle deflection. The effective-dihedral param-eter Clp, the damping-in-roll parameter CLp, and damping-in-yaw parameter Cnr allincreased (became more negative) with increasing trim normal-force coefficient. For thelatter three parameters, the ef
8、fect of nozzle deflection was dependent on the trim normal-force coefficient.17. Key Words (Suggested by Author(s) 18. Distribution StatementParameter extraction Unclassified - UnlimitedMaximum likelihoodAerodynamic coefficients- STAR Category 0219. Security Classif. (of this report) 20. Security Cl
9、assif. (of this page) 21. No. of Pages 22. Price*Unclassified Unclassified -2 -s $3.25For sale by the National Technical Information Service, Springfield, Virginia 22151TProvided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-LATERAL STATIC AND DYNAMIC AEROD
10、YNAMIC PARAMETERSOF THE KESTREL AIRCRAFT (XV-6A)EXTRACTED FROM FLIGHT DATABy William T. Suit and James L. WilliamsLangley Research CenterSUMMARYA maximum likelihood extraction method has been used to extract the lateral staticand dynamic aerodynamic parameters of the Kestrel aircraft from flight tes
11、t data. Theaircraft configurations included thrust-jet angles of 00, 150, and 300, and the test Machnumbers were 0.43, 0.62, and 0.82. The results showed that most of the parametersvaried linearly with trim normal-force coefficient. The directional stability parameterCno showed a small increase with
12、 increasing trim normal-force coefficient and also withnozzle deflection. The effective-dihedral parameter Cl0, the damping-in-roll parameterClp, and damping-in-yaw parameter Cnr all increased (became more negative) withincreasing trim normal-force coefficient. For the latter three parameters, the e
13、ffect ofnozzle deflection was dependent on the trim normal-force coefficient.INTRODUCTIONAnalytical and simulator studies of the flight and handling qualities of aircraftrequire that accurate estimates of the aerodynamic parameters be used if the results areto be valid. There are several methods of
14、obtaining the parameters. These includemethods presented in various books, wind-tunnel tests, and extraction of derivativesfrom flight test data. Of these methods, derivatives determined from flight tests shouldbe the most accurate since the results are obtained with the aircraft in its properenviro
15、nment.To provide aerodynamic parameters that were difficult to determine by othermethods, parameters have been extracted from flight data for many years. In the past,many of the attempts yielded unacceptable results. At present, improvements inProvided by IHSNot for ResaleNo reproduction or networki
16、ng permitted without license from IHS-,-,-instrumentation and particularly the development of large capacity high-speed computersenable the engineer to take advantage of the advanced mathematical methods of parameterextraction. Example results from recent studies made at the Langley Research Centerb
17、y using an advanced extraction method are reported in references 1 to 4.The purpose of the present study is to determine the lateral aerodynamic param-eters of the Kestrel aircraft from flight data for several airspeeds and thrust vectorangles. The technique and program used in extracting the parame
18、ters is that ofreference 5.SYMBOLSValues are given in both SI and U. S. Customary Units. The measurements andcalculations were made in U. S. Customary Units.a acceleration, m/sec2 (ft/sec2)b wing span, m (ft)Fj primary engine thrust, N (lb)Fy aerodynamic forces along aircraft Y-axis, N (lb)FZ aerody
19、namic forces along aircraft Z-axis, N (lb)g acceleration due to gravity, Ig = 9.81 m/sec2 (32.2 ft/sec2)h altitude, m (ft)I moment of inertia, kg-m2 (slug-ft2)M Mach numberMX moment about X body axis, N-m (ft-lb)2Provided by IHSNot for ResaleNo reproduction or networking permitted without license fr
20、om IHS-,-,-MX, j rolling moment due to reaction jets, N-m (ft-lb)MZ moment about Z body axis, N-m (ft-lb)MZ, j yawing moment due to reaction jets, N-m (ft-lb)Nf engine fan speed, percent of maximum speedp rate of roll, rad/sec or deg/secq rate of pitch, rad/secq dynamic pressure, pV2, Pa (lbf/ft2)r
21、rate of yaw, rad/sec or deg/secS total wing area, m2 (ft2)u velocity along X body axis, m/sec (ft/sec)V aircraft total velocity, m/sec (ft/sec)v velocity along Y body axis, m/sec (ft/sec)W aircraft weight, N (lb)w velocity along Z body axis, m/sec (ft/sec)Xi individual state in complete state vector
22、 Xa angle of attack, rad or degP angle of sideslip, rad6a differential aileron deflection (positive for right aileron up), rad or deg3Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-6r rudder deflection (positive for trailing edge right), rad or deg0
23、 pitch angle, radej jet nozzle angle, degp air density, kg/m3 (slugs/ft3)0 roll angle, radMXCl rolling-moment coefficient, 11 pV2 SbMZCn yawing-moment coefficient, 12 pV2 SbFyCy side-force coefficient, 1 FpV2 SCZ normal-force coefficient, F ZaCCCp = pba2VaC/r rb2VaC/aC,416 a4Provided by IHSNot for R
24、esaleNo reproduction or networking permitted without license from IHS-,-,-C16r a-raCnC%- pb2VaCnCnr rb2VaCnCpCng apCn aC6a 86a8CnC-,Cn5r a6rC CyCYP pbaCyCYr 8rb2VBCyCY0 a_CyCY6r - rSubscripts:c computed5Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,
25、-m measuredo indicates coefficient at trim conditionst indicates state at trim conditionsX X-axisY Y-axisZ Z-axisA dot over a symbol signifies a derivative with respect to time.The following symbols are used only in figures 4 to 6 and result from the limitationsof the computer-controlled plotting eq
26、uipment:AYI lateral acceleration, m/sec2 (ft/sec2)DA =6a - a, t, radDR = r - 5r, t, radP roll rate, rad/secPHI roll attitude, radR yaw rate, rad/secV lateral velocity, m/sec (ft/sec)6Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-EQUATIONS OF MOTION
27、The equations of motion used in this study are referred to a body-axis system andare as follows:Y-direction:pw - ru + g cos 8 sin5 + pV2 S - C Y , + CY + Cyp + CYr+ CY6 (r -6r, t) (1)Rolling moment:IXZ I - IZ IXZ MX, j 1 pV2 Sb rr P+ X-p+ IX Cl,o+Clpb rb (2)+ Clp + Clr 2V + Clr (br- r, t) + Cla (6a
28、- 6a, t(2)Yawing moment:IXZ IX - XZ Mzj 1 pV2Sb +C pbr= + Z qr + + 2 Iz o + C- ,+ I I IZ Inr np 2V+ Cnr b + C n( r - 6r,t) + Cn (6a -ba,t) (3)TEST AIRCRAFT AND EQUIPMENTThe test aircraft used in this flight investigation was a Hawker Siddeley Kestrel(XV-6A). The Kestrel is a single-place, prototype,
29、 vectored-thrust, V/STOL strike-reconnaissance aircraft. A three-view drawing of the aircraft is shown as figure 1.A single Rolls-Royce Bristol Pegasus 5 turbofan engine powers the Kestrel. ThePegasus is an axial-flow, vectored-exhaust turbofan engine, with a 1.4 bypass ratio,having an uninstalled s
30、ea-level static thrust rating of 68 900 N (15 500 lb). Thrust is7Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-vectored through two pairs of controllable engine exhaust nozzles (shown in fig. 1) and isequally distributed between the forward nozzles
31、 which exhaust bypass air from the fanand the aft nozzles which exhaust turbine air. The nozzles are mechanically intercon-nected and can be rotated at rates up to 90 deg/sec to any position from fully aft (0j = 00)to 50 forward of vertically downward (Oj = 950). Nozzle angle is controlled by a sing
32、lelever which is located inboard on the throttle quadrant and which is the only additionalcontrol required for thrust vectoring in the Kestrel.Control moments during nonvectored flight are provided by conventional aerodynamicsurfaces. The ailerons and tail plane are powered by tandem hydraulic syste
33、ms; the,rudder is unpowered.During vectored flight, reaction control moments are added to those produced bythe normal aerodynamic surfaces. Reaction control shutter valves, located at the nose,tail, and wing tips, are mechanically connected to their corresponding aerodynamiccontrol surface and recei
34、ve high-pressure engine bleed air as a function of engine nozzleangle. (See fig. 1.) Full reaction control is provided at engine nozzle angles greaterthan 300. No stability augmentation system (SAS) is provided. However, during flight atlow dynamic pressures where the pilot does not get feedback to
35、the control stick fromforces on the control surfaces, an artificial-feel system is provided. Lateral feel isprovided by a nonlinear spring unit and longitudinal forces are provided by a g-feel unitsupplemented with a feel spring. This g-feel unit is a bobweight in the control run whichincreases long
36、itudinal maneuvering forces by 8.9 N/g (2 lb/g) for normal acceleration and4.9 radN r e.1 lb c for pitch acceleration.rad/sec2 Iradsec2The rolling and yawing moments due to the reaction jets are given by the followingempirical equations which were obtained from the manufacturer:M -2.1 4 1Nf M12 X0 ,
37、Mz, j= 1-2.14 -I Mz, jwhere I1= for 0 208Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-and where MX, j and MZ, j are taken from figures 2 and 3, respectively. Theinformation for figures 2 and 3 was taken from a manufacturers report at the time ofth
38、e problem setup. Modifications to the aircraft probably have resulted in a differentreaction jet curve. Since the reaction jets are not particularly effective (in comparisonwith the aerodynamic controls) over the test Mach number range, any differences thatexist were not considered significant. Addi
39、tional aircraft and engine data are presentedin table I.FLIGHT TESTSThe aircraft was flown at nominal Mach numbers of 0.82, 0.62, and 0.43 with threenozzle deflections, 00, 150, and 300, and at constant thrust for each flight condition. Thealtitude for the flights was about 4.6 km (15 000 ft). At ea
40、ch nozzle deflection and air-speed several runs were made with the pilot perturbing the aircraft from a Ig levelflight condition with different combinations of aileron and rudder inputs. The pilots wereinstructed to exercise the rudder and ailerons by means of successive inputs in both thepositive a
41、nd negative senses. The form of the inputs was generally left to the pilot buthe was instructed that, if possible, home of the inputs should be abrupt. Flight testconditions are given in table II.For the purpose of this analysis, the center of gravity was assumed to be at13.7 percent wing mean aerod
42、ynamic chord, and the average values of weights and inertiaslisted in table I were used. Data pertinent to this study were recorded during flight tests.The recorded quantities which were used are listed in table III. The full-scale range ofthe recording instruments and their response frequency are a
43、lso given in table III. Afilter was used to limit the response frequency of the instruments. In general, the instru-ments had responses which were flat past the filter frequency. The accuracy of thesemeasurements was estimated to be 3 percent on full scale for the states measured. Thea and p vanes u
44、sed in this investigation were located on a nose boom approximately1.8 m (6 ft) in front of the aircraft. The vanes were made of balsa wood and the errorintroduced by the vanes themselves was small compared to the estimated accuracy of3 percent of full scale. The readings from the a and P vanes were
45、 corrected for theeffects of angular rates.9Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHS-,-,-PARAMETER-EXTRACTION TECHNIQUEAll the data were stored on an onboard magnetic tape recorder by using wide-bandFM recording techniques. The linear velocities a
46、long the vehicle body axes were calcu-lated from the measured airspeed, angle of attack, and angle of sideslip. The correctedand calculated data were put on a tape for use in the parameter-extraction program at arate of 20 points per second. Additional details on preparation of flight data for theex
47、traction program are given in reference 1.The parameter-estimation procedure used in this study is an iterative procedurewhich maximizes the conditional likelihood function L. This L is a function of theaerodynamic parameters, weights, and initial conditions given by the relation)1/2 1/2 exp -im- Xi
48、) R1 Xi)m- (Xi cwhere N is the number of data points, R is the estimate of the error covariancematrix, (-i)m is a vector quantity whose elements are the measured states, and (Xi)c isa vector of the calculated states whose elements are determined from the integration ofequations (1) to (3). Maximizing the likelihood function minimizes the difference betweenthe measured and calculated aircraft motions (Xi)m- (Xi)c. A detailed description ofthe method used is given in reference 5.The states used in the likelihood function were v, p, r, and ay. In this studythe in
