OPTIMIZATION Davidc T.

AND

PERFORMANCE

ESTIMATrION

Johnson Dynamics Laboratory Air Force Base, Ohio

Air' Force Flight Wr'ight-Patterson November 1972

DISTRIBUTED BY:

National Technical Information Service U.S.DEPARTMENT OF COMMERCE

5285 Port Royal Road, Springfield Va. 22151

La

AFFDL-TR-72-58

0 4ttq

EVALUATION OF ENERGY MANEUVERABILITY PROCEDURES IN AIRCRAFT FLIGHT PATH OPTIMIZATION AND PERFORMANCE ESTIMATION

DAVID T. JOHINSON

TECHNICAL REPORT AFFDL-TR-72-58

DDC'
NOVEMBER 1972 rflI"FP P

l. Ir,

B
Approved for public release; distribution unlimited.
R.p,.od , ed by

NATIONAL TECHNICAL INFORMATiON SERVICE

L

S DePorIenlt of Co.

er.c

Spigfield VA 22151

AIR FORCE FLIGHT DYNAMICS LABORATORYAIR FORCE SYSTFMS COMMAND WRIGI-CH-T-A1TTERSON AIR FORCE BASE, OHIO 45433

V4d

NOTICE

When Government drawings, specifications, or other data are used for any purpose other than In connection with a definitely related Government procurement operation, the United States Government thereby incurs no responsibility nor any obligation whatsoever; and the fact that the government may have formulated, furnished, or In any way supplied the said drawings, specifications, or other data, is not to Wo regarded by implication or otherwise as in any manner licensing the holder or any other person or corporation, or conveying any rights or permission to niangafacture, use, or sell any patented invention that may in any way be related thereto.

A!
NTI

'i,!~i. AV-AI1LA8iLITy Ct3DES

Copies of this report should not be returned unless return As required by security considerations, contractual obligations, or notice on a specific docunoent.
AIR FORCE/56"80/1B January 1973 100

I I

II

UNCLASSIFIED
S -,ntr Cla 1sSftcatlon'

(Se-ret)" 'a~sif~ti'tor
OiiIGIr ATING

if

fil,

~tl

D.OCUMENT CONTROL PATA

aitraci ad it.dening At0'iotntionn-r~

R& D
be siiretd write
ihe

bid)thor) i

uverall reprt! I.. clsesilled)

'V C:.ASSIFICA.ION

-IViro (C01ro7

i.t1t

Air Force Flight Dynamics Laboratory A FFDL/FXG Wright-Patterson Air Force Base, Ohio 45433
RI'ORI E "TITLE

LPOR"

StICUVI

UNCLASSIFI2WD

2h. GOUP

EVALUATION OF ENERGY MANEUVERABILITY PROCEDURES IN AIBCRAU FT FLIGHT PATHI OPTIMIZ1ATION AND PERFORMANCE ESTIMATION A OCSCRIPTIVL ENOTS (7ype o riper;r and incf.sve due.s) Final Technical Report July 1967 to August 1970
*AU TUG SCSI (bitr

la

,me.,

iilddie

initial, lent flame,)

David T. Johnson
O REPORT DATE '1, TOTAI NO. OF PAGES

T.

ri).

OP

RES

November 1972
8a. CONTRACT OR GRANT NO

87
QA. ORIGINATOR'S RIEPONT NIMER[;

b. PROJECT

NO.

1431 143109
I'lb

AFFDL-TR-72-53
OTH"R REPORT NO(S)

Task No.
d.
i0 DISTRISUTIOJN

thI rapnoen)

(A

IV Ot,-S

,.a.;may

A.aSlWaed

STATEMENT

Apnproved for public relea-e; distribution unlimited.

II SUPPLEMENTARY NOTES 12. SPONSORING MILITARY ACTIVITY

a!

Air Force Flight Dynamics Laboratory WrIght-Patterson Air Force Base, Ohio 45433

An assessment is made of the applicability of Energy Maneuverability techniques (EM) to flight path optimization. A series of minimum time and fuel maneuvers using the F-4C aircraft were established to progressively violate the assumptions inherent in the EM program and comparisons were made with the Air Force Flight Dynamics Laboratory's (AFFDL) Three-Degree-of-Freedom Trajectory Optimization Program and a point mass option of the Six-Degree-of-Freedom flight nnth program. t was found the EM results were always optimistic in the value of the payoff functions with the optimism increasing as the percentage of the maneuver involving consttant energy transitions Increases. For the minimum time paths the resulting optimism was less than 27%1 for the maneuvers where the constant energy percentage was less than 35.',", followed by a rather steeply rising curve approaching in the limit 100% error for paths which are comprised entirely of constant energy transitions. Two new extensions are developed in the report; the first is a varying throitle techni(ue for use on minimum fuel paths and the second a turning analysis that can be applied in conjunction with a Rutowski path. Both extensions were applied to F-4C maneuvers in conjtnction with 'Rutowskils paths generated from the Air Force Aimament Laboratory's Energy Maneuverability program. The study findings are that energy methods offer a tool especially useful in the early stages of preliminary design and functional performance studies where rapid results with reasonable accuracy are adequate. If the analyst uses good judgement in its applications to maneuvers the results provide a good qualitative insight for comparative purposes. The paths should not, however, be used as a source of maneuver design or flight schedule without verification especially on relatively dynamic maneuvers where the accuracy and optimality of the method decreases.

DDFORM1 4

73

UNCLASSIFIED
SecUrity Classifi'nuop

S,

vL! tIs It Ii nil,-atitW

LINK A,

LIN ROLL

LINK

C W

RcLC

WT

WT

ROLE

A Lva .ft
1L

Iliglit P~ath olitimin zucvraii

ation

I g;

Applications

.lininiiuiin Fuel/Tinic

oprsm

I~

fic.1972 iti9

7594463

9,:9,

UNCIASSIFIED
Secu~rity

CosIficstmkt~n

EVALUATION OF ENERGY MANEUVERABILITY PROCEDURES IN AIRCRAFT FLIGHT PATH OPriMIZATION AND PERFORMANCE ESTIMATION

DA VID T. JOHNSON

Approved for public release; dis.ibution unlimited.

I/I

'P'h>

et

\,a.

'cpa red by peresonnel of the High Speed Aero Pcrtormance )ivisioi uf the Air Force Flight Dynamics Laboratory The rescarch was

Sr>.l,, lligIht IMechanjcis

(A\In..EI"G), \Vritgh- Patterson Air Force Base, Ohio, , , II I 01 1Millia

I1C j'v

UI k I HL

lrtject 1-1:3j

"' Pliht Meceknies aid Pc rfoimanco 1;valuationi

Vchich's", and Task 1.-1 109 "F'light Path Analysis Methodology for

CumI)at Aircraft and Missiles".

The work period extended from July 1967 to August 1970.

Acluiowledgement

ij mna.de of (he help received from B. 11, Benson for overall guiduance and R. C. Niash fOr the optimized flight paths run on AFFI)L's Three-.Degree-of-

1"I'Oudom Optimization Program.

This report was submitled by the author May 1972.

'Itc]

nijeal veprt ha.een reviewed and is approved.

IP. ANT'I)NATOS Chic, lFlit ilMtchanics 1)i\ ::on Al' Flight DyInaics Laboratry

iv

TAB3LE OF CONTENTS

SECT'1ON INTRODUCTION ME'I'O1IS 1. Energy Maneuverability Mlethod's a.

b. 2.

,AE

N 3
4

Specific Excess Power

flutowski Paths

Of Energy, Climb a. b.
c.

Limit:ttions and Possible Extensions

4

Thrust Along Flight Path Path Fllyability and Optimality

Constant Throttle Setting

d.
III

Vertical Plane Restriction

I;

VERTICAL PLANE MANEUVERS 1. Minimum Time Paths a. U. c. Flyalility and Accuracy Optiniality Discussion
I

9 9

*5

AMiniltuel a. h., c. Throttle Switching Maneuver Results Discussion

-10SPaths

aa 3s

Zooni lianeuver

TABLE' OF CONTE'NTS (CON TD)

"EICTiON
lV OUT OF -PLANE MANEUVERS I. Minifimuml T Im1e Paths a. Trn-ing Char11ts
13.Fyahility and Accuracy
C.

PA GE 42 42 42 51 56 6

Opltillality Discussion

d. 2.

M-inlirnuI1 Fulll Paths a. b. Turning Charts Maneuver Res,,ults

61 () 1 63 71 73

CONCLUSIONS

liE FE RELNC ES1

vi

I LLUSTRA TION6

FIGUHE 1. 2.

TITLE Specific Excess Powcr (P s) Contour Map Category 1, Maneuver 1, Minimum Time Path-EM and SDF Category 1, Mancuver 2, Minimum Time Path-EM and SDF Category 1, Maneuver 3, Minimum Time Path-EM and SDU Category I, Maneuver 3, Modified Minimum Time Energy Path Category 1, Maneuver 1, Minimum Time Path - EM and TOP Category I, Maneuver 2, Mininmum Time 1'ath-EM and TOP Co-tegory I, Maneuver 3, Minimum Time Path-EM]l and TOP Category 1, Short Minimum Time Path-EM_ and TOP Percent lkviation in E M Results-Category I, Minimum Ti me Maneuvers Maneuvering Energy (Lr) Contour Map-Maximum A/13 Power Maneuvering Energy (E .
SA//-[

PAGE 12

15 16

3.

4.

18

5.

20

6.

22.

7.

23

8. pI 9.

24

26

10. 11.

27

29
30

12. 1:3.

Power

Contour Map!-Minimmn Contour Map-Military

Maneuvering Energy (Q') Power

31 32

1,4. 15.

Throttle Efficiencies Along Minimum Fuel Paths lRegions for 1est Fuel Efficiency-3 Throttle Analysis Category 1, Maneuver 1, Minimum -ucl Path-EM and SDF Category I, Maneuver 2, Minimum Fuel Path-ELM and SDF

34

16G.

35

17.

36

vii

L"II

,1lPSTBA'l'IONS (CONITINUED)

FIGIURE I1.
19.

TITLEl Category 1, Maneuver 3, Minimum Fuel P1ath-E 1 and SDl'

Zoom Maneuver to Maximize Altitude

PAGE 37
40

20o. 21,
22.

lorizontal Turning Rates Acceleration Time Penalties Along Minimum Time Paths
Turning Efficiency Along Minimum Time Paths

43 46
48

23. 2-1. 25. 26. 27. 2$.

11, Maneuver 1, Minimun Category T furning lEfficiency

Time

50

Category II, Maneuver 1, Minimum Time-Bank Schedule Family Category II, Maneuver 1, Mlinimum Time Path-L M and SIF Category II, Maneuver 2, Minimum Time-Bank Schedule Family Category I1, Maneuver 2, Minimum Time Path - EM and SDF Category 11, and TOP Ml\aneuver 1, Minimum Time Path -EM

52 53 54 55 57 58

"29.
30. 31. :32. 33. 34. 35. :86;.

Category 11, Maneuver 1, Minimum Time-EM TIn.' T01, Bank Angles

S. .... ..'..

Category 11, Maneuver 2, M\hinimumn Time lPath-EM Category 11, Maneuver 2, "Turn,/TO1 Pbink Angles Minimum Timc-EM

59

60 64 65 68 69 70

Acceleration Fuel Penalties Along Minimum Fuel Paths Turning Efficiency Along Minimum Fuel Paths Category II, Maneuver 1, Minimum Fuel Path-EM and SDI)I' 13ank Schedule for the Minimum Fuel Maneuvers Category II, Maneuver 2, Minimum Fuel Path-FAl and S1)F

"viii

TAABLE PA GE
-

I II III
iNV

Category3 I

Vertical Plau1C Mancux-crs,

-

10

( a fe g O r y I M \a ne u ve r' I Staging Example

tatgorA,

I V~ l nij nijj u

Fu 14

OUt Of

P~lane Maneuvers

43

V VI
N'II

M\Iiniu)Lm Tlime Path~ Turnin~g Efficiency Example Mlk~inumn Time Constant Efficiency Turn Example, Minijmumn Fuel Path Turninig Efficiency E-xajniple. Mii\LflLfl Fuel Constint Eficiency Tuirn Example

45. 49 62 66

ix

SYMBOLS

C Llift D E 1F1 Es g 11 K 1 M n, N m Ps q Ta 'lX, Ty t TE V Vavg

coefficient drag total energy maneuvering energy specific energy acceleration of gravity altitude constant used in section III lift Mach number normal load factor ma. 3 specific excess power dynamic pressure thrust available thrust components in wind axis system time turning efficiency velocity average velocity weight

wVa
Wf

fuel weight available for use fuel weight angle of attack

AX

variable X increment for given change in variable y where x and y can he any two variables

SYMBOLS (CONTINUED) flight path angle

X
96
qi

engine cant angle bank angle heading angle

subscripts f i, o p final condition initial condition penalty

dot over symbol indicates differentiations with respect to time

xi

SECTION I INTRODUCTION

In maling a comparison of two aircraft, in evaluating a design, in determining suitable operational marneuvers, ald in measuring the performance of an a ir !raft, optimum flight paths are invaluable as a standard. One of the more

accurate ways of determining an optimum, path is through the use of a program such as the Air Force Flight Dynamics Laboratory's (AFFDL) Three-Dimensional Trajectoi y Optimization Program (TOP) (Reference 1). The TOP program

uses the method o& steepest descent which is an iterative scheme starting with any nonoptimal psth and deriving an improved trajectory in each iteration until the payoff function is optimized and all constraints are satisfied. This program

provides a comr lete time history of the optimum path in three dimensions, incorporates complete equations of motion, includes realistic vehicle representation, and can use any integrated variable as a payoff quantity. The dis-

advantage of the program il that it takes a relatively long time to reach an optimum due to the repeated solution of the equations of motion for each path improve me nt.

A second technique using the methods of E. S. Rutowski (Reference 2) is based on the use of total energy to provide an approximation to solutions of minimum time or minimum fuel problems. The technique has been extended The method is predicated on Ig

and developed into a computer program called Energy Maneuverability (EM) by the Air Proving Ground Center (Reference 3).

level fliiht and lends no insight into sunh pnrameters as load factor or pitch angle along the path. However, a solution is obtained very rapidly making the

technioue attractive where rapid solution time is necessary. The objective of this study was to determine the applicability of the EM techniques to flight path optimization and to investigate improvements to alleviate some of its restrictions. Section lI contains the basic equations used in

the study and a discussion of the assumptions and limitations of the EM method. Two new extensions are developed, the first a varying throttle technique for use in generating minimum fuel paths and the second a turning analysis that can be applied in conjunction with a Rutowski path. Both the varying throttle and the

turning analysis were designed so they can be applied using the existing EM program and a desk calculator although the methods could also be fully automated. A series of minimum time and fuel maneuvers was then esctablished using the F-4C aircraft and comparisons made with AFFDL's TOP program and a point nuass option of the Six-Degree-of-Freedom (SDF) flight path program (Reference 4). For the minimum fuel paths both maximum power and varying throttle The below table indicates the type of comparisons

paths are used; for the out of plane maneuvers results both with and without the turning analysis are shown. made, maneuvers selected, and programs used in Sections III and IV SECTION MANEUVERS Baseline III Vertical Plane Out of plane EM and modified EM modified EM METHODS Flyability/accuracy Comparison SDF SDF Optimality Comparison TOP TOP

IV

SECTION II METHODS

1.

ENERGY MANEUVERABILITY METHODS The basic energy maneuverability equations used in this study are outlined

in the following paragraphs.

A more complete description of energy maneuver-

ability methods may be found in References 2, 3, or 5. a. Specific Excess Power

The energy state of a vehicle is defined as the surn of its potential and kinetic energy: E and specific energy:
=

W (h + V 2/2g)

(1)

2|
Es zE/W =(h + V2/2g) the time derivative of Es is then: (2)

= h+ V V/g
fliaht

Writing tile equation of motion along the paraineters as follows: W/g (Ta T

path,

is related tuov-..

D -

sin W+ sn/g (5)

aD)/W sin

V (Ta-

D)/W

V sin- + V V/g

+ V

V/g

(6)

specific excess power S then: P. = Es = V(Ta -D)/W Computation of Ps at a given h, V, and n is performed by setting: L = nW (for level unbanked flight n = 1) Ta and CD are obtained from tabular listings of aircraft characteristics at the specified conditions and Ps computed from Equation (7). Contours of Ps

(7)

may be obtained by repeated solution and interpolations over a mesh of points. b. Butowski Paths

Computation of Rutowski paths is based on minimization of the integral: "

rS I

(9) ([I/dE,9/dt)dE$

for a minimum time path from a lower to a higher energy stale and

"Wf

ES=

[l/d Es/dwf] dEs

(10)

for a ninimum fuel path. Th:., method of finding the Mach - altitude history for the minimum time paTd Ca,. be shown graphically as the points of tangency between the P and E contours. Similarly the Mach - altitude history of the minimum fuel path can be shown as the points of tangency between the Ps / and Es contours. For

specific maneuvers initial and fir td conditions off the flutowski paths are reached by ,vonsLintenergy dives or zooms. r-utowski paths for selected maneuvers will be shown in Section III.

2.

LIMITATIONS AND POSSIBLE EXTENSIONS OF ENERGY CLIMB a. Thrust Along Flight Path

The usual assumption is that the thrust is directed along the flight path as in Equation (4) when in fact its direction is a function of angle of attack and cnghic cant angle. follows: A change to include these effects could be incorporated as

Breakinrg the thrust into components Equation (7) becomes, Ps

VA/ (T-D) where T,
=

cos (a +X)

(II)

The corrected lift distribution, in place of Equation (S), would then become, L=nW - Ty where Ty =Ta sin (a +X) (12)

or ClqS = n" - T 8 sin (aA)X)

(13)

With tabular aerodynamic coefficient input, both sides of Ec aatlon (13) are a function of a (since C at a parti'ular M, li. varies with a ) and the equation may be iterated for a

qnd

n.

would then be ,,-t-

-'qu-ti(,-

I-.. by

(ii),

As this change in computational procedure would increase running time and complexity of the 1CM program it was not implemented in this study. b. Path 1'iyability and Optimalit,

llotowski paths are computed at constant load factors and there is no assurance that tho resultant path is flyable as the full dynamics involved airo not considered. For example a result of the energy method is that energy can be convcrted from potential to kinetic or vice versa along lines of constant energy

:n zero time with zero fuel expended.

This is physically not possible so it is

ne(:essary to deviate from the path in practice producing a difference in the fin-al result. A inajor portion of this study effort has been devoted to assessing In the next two sections smoothed

the flyability and optiniality of Butowski paths.

flight path angle histories from Rutowski paths for specific maneuvers were input to A'I,'DL's trajectory program (Reference 4) to assess the flyability and realism of the energy generated path. Minimum time paths were then generated

o' A*FI'DL's trajectory optimization program (Reference 1) to compare with the above two results.

The theory for the Putowski path computation is defined for increasing specit.ic energy only, hence, no decreasing energy maneuvers were considered. llhowever, the investigation did include a zoom maneuver as will be shown in the next section. References 3 and 6 suggest the use of a "rule of thumb" to develop

corrections to the enaergy path along the constant energy segments to improve flyability of the paths. This approach was applied to a sample maneuver in the

next. se: tion,

.
uii 1,

-tUlL

I"

,roilt-le Stcting

HiSttorically lluhtwski paths have been computed at a constant throttle settiKg lor both minimum time and fuel. To investigate the gains possible by vary-

ing the throttle for minimume fuel paths, three throttle settings were used in this

Msd' , military, minimum afterburner (A/B), nnd nmximum A/B.

Switching

between throttle settings, and corresponding Btutowski minimum fuel paths, \was accormplished to minimize the integral of Equation (10).
are shown in Section III
-

Spc.zific examples

2a.

d.

Vertical Plane Restriction

The energy climb procedure considers only the changes in energy state )f a vehicle and hence does not permit turning maneuvers. However in practice many of the maneuvers of interest to analysts and pilots include varying amounts of turn as well as changes in the vehicles energy state. The following turning
procedure was used in this study to obtain an estimate of turning performance in conjunction with a Rutowski path. This method was designed to permit

cumnputation ef turning mancuvers by any person who has a niethod of computingg 11uto~wski p.iths such an the program of Reference 3. With the assumption of level flight and that the thrust is directed along the flight path we can write using Equation (8):
L cos
=

(1G)

and

cose

(15)

writing the equation in the lateral direction "niV q L s h, nW

Vr

1(t6)

or~

g/v(

2)(7

Equation (17) was then applied between energy levels along consLant load factor Rutowski paths to obtain . The amotut of turn between energy levels at a given load factor was then calculated by:

^A.-. q" A.T

where A t is the time increment to go between energy levels along a Butowski path calculated by Equations (7), (8), and (9). levels by: The time penalty ( A t) to operate at a load factor higher than one is then calculated between two cnergy

At

At(n > 1) -

At(n = 1).

(1!i)

Tie turning efficiency between energy levels was then defiled as:

(Tl)

A q,/ A t

for n > 1 for n > 1

for n = 1.

on mhnimuni

time paths (20)

(TA) = A
(TE)=0

/A

W 1

on minimum fuel paths

The turns used in this study were computed at constant values of turning efficiency (TV) and hence are called "constant efficiency turns" in the following sections. Specific examples of the calculation procedure are shown in Section IV.

SECTION Ill VEETICAL PLANE MANEUVERS

The maneuver t.ype and the initial and final conditions for the nonturning maneuvers selected are shown in colunms 1, 2, and 3 of Table 1. Maneuver I is a long climb acceleration with initial and final conditions close to the llutowski path. energy level. Maneuver 2 is a shorter climb acceleration with initial condiManeuver 3 has both initial and final conditions at the saule alColumn 4 shows the method used, tions considerably away from the Rutowski path in the h-M plane and at a higher titude starting at a higher and ending at a lower energy level than Maneuver 2. Maneuver 4 is a zoom to maximize altitude. cussed in the following paragraphs. colunu 5 the tinme required, and colunm 6 the fuel expended; these will be disData for the F-4C aircraft used in the simulations was obtained from the manufacturer. 1. MINIMUM TIME PATHS a. Flyability and Accuracy

The 1utowci M minimum time paths were run for the above maneuvers using the program described in Reference 5 at a 1 g load factor. Figure 1 shows the P contours and the energy path resulting from connecting the points of tangency between the P and E contours for the F-40 at maximum A/B power. The Rutowski path changes only slightly with small changes in aircraft weight, thus

all of the maneuvers basically are made up of segments of the path shown
sired initial and final points.

in[

Figure 1 together with constant energy segments to connect the path to the deThe maneuver results are shown in Table I as "EM".

To investigate the flyability and accuracy of these EM paths the maneuvers were simulated on option 6 (point mass) of the SDF program (Reference 4).
certain amount of modification to the path was required for simulation on the

SDF program as far as smoothing the path history.

Since constant energy dives,

and zooms are not achievable in practice (without setting Ta = D) the pushovers were done at low positiv~e lift and the pull-ups at moderate positive load factor.

.cl1

IT 002r b

CV In

C-70 en ' tfln LA

CA~ 71r, O L '. %Qa Nmo

C0 -

0-7H ( 'N- r1O -m CAo o t Nr O..~.t)

0% P-. '

r4 IT 0 0 00ir 'c C'C''

' C j-; NM

Ca' qm

00 ch mt
vi LnNeq-

0 r, N q

Oe4au (7 D rJco

(NLn %010

Af

N t

-JJ

*
as/

9w Wu L

NN

I-.

wEr

p4n

Liwam47

jbdJ~
i

C)

C) 00

tMt 4-,0

.00

100

-0

C00 0 c
-

00
Q Q\

00

4) u

>

x:

X 0

CI
0

C:

-J
C--

CL t

0f

00 C, 0

tD
c;.

CDC

CLC

0C

.LlI Pn4!4IV

Two flight plan programmers were used in the SDF program n

=

f (11) and

f (M).

The flight path angle history was calculated from the EM history by:

sin1

(A h At V)

(21)
The load factor programmer was used Staging of controis

and used to coiftrol, most of the path.

primarily to perform climbs at near constant Mach number.

was used at various points along the paths to change modes of operation. Table I! shows a typical example of the staging and control used in SDF simulation. The example consists of a level acceleration at military power, a subThis path was split into nine control segAn

sonic climb at minimum A/B, a pushover/pullout transition at maximum A/B and a supersonic climb/acceleration.

ments or stages as shown in the Table II to facilitate the SDF simulation. SDF paths are listed in Table I as "EM-SDF".

initial flight- path angle of 00 and unconstrained terminal flight path angle was used,

Figure 2 shows the paths for the two computation methods for Maneuver 1 which has starting and ending conditions near the Rutowski path. The figure shows the SDF and EM times are close all along the path with the principle difference in the path profiles occuring in the transonic area where the EM path has a constont energy dive due to the shape of the Ps contours shown in Figure 1. The agreement between methods is quite good with the EM being 4.3 seconds or 1. 3%,) optimistic in time and 49 pounds or 0.9% optimistic in fuel. Fig-ure 3 shows the pathQ fnr
t

.aneuwer 2 which uses the same end condi-

tions as Maneuver 1, however, the initial conditions are at a higher energy level and are not close to the Rutowski path. transition segment. To reach it requires a sizable The principle difference in the path profiles occurs in the 1.25). The SDF times lag the EM times by a wider On this path the agreement is Agreement

initial constant onergy dive (. 7 < MI < 0.95) and in the high transonic -supersonic area (0.95 < M ( not quite as good.

range than in Maneuver 1 over the entire path.

The EM being 10 seconds or 3.7% optimistic.

in fuel used is quite good (7 pounds or 0. 2%) apparently due to the SDF path being higher over the diving portion saving fuel which is later consumed by the longer time required to reach the final conditions.

13

TABLE CATEGORY I, MANEUVER

II

I - MINIMUM FUEL STAGING EXAMPLE

Stage No

Staging Parameter to next stage

Controls

Remarks

Power Setting

I 2 3 4 h

=
-

.60 17,000 ft -y n n =
=

0 f(h) f(M) f(M)

level flight climb/accele-rate clInt/accelerate constant Maci c11imb Pushover at low lift Pullout and Accelerat ion/ climb rapid climb at near constant

Military Min A/B iMn A/B Min A/B

y
h
=

18'
40,500

M -

1.21

f(h)

Max A/B

'

1.84

f(hi)

Max A/B

47,200

f(h)

Max A/B

Mich
8 M = 1.913 "
= .1

near level accelerating flight Pushover and dive to end conditions

Max A/B

M =

f(h)

Max A/B

14

eI ! II II I I I

I ii i ii

-.

(V

0~I

0
4A

rs.4
E
I

~0

Nob

100

oo
0

ORA,-

01)9
6 0
S

40

4-.

CL
Vh ~) w4

0 N

(
all

ap4!I
150

00 0o

4L

(1

-4
p1 4 1 N 'e 0

00

NO

N,

EL

a.
0

W a

U)
9-.-10

:L
*

/m

/A3nai

Maneuver 3, Figure 4, has both starting and ending conditions at 35, 000 feet. The path starts at a higher and ends at a lower energy level than Again, the initial conditions are well off the Rutowski path. This region The final The Maneuver 2.

EM path has a long initial constant energy dive which does not connect to the subsonic but to the supersonic portion of the Rutowski path. (0. 75 < M < 1. 3) accounts for the major difference in the paths and the SDF lags the EM times by a larger amount than in the other maneuvers. pounds or 0. 5% optimistic ;n fuel. The indications from these three maneuvers are that the EM procedure predicts times with reasonable accuracy ranging from less than 2Vc optimistic for a 310 + second path with initial and final conditions close to the path to about 8% optimistic for a 130+ second path with initial conditions well off the path. The associated fuel required for the maneuvers showed surprisingly good agreement, being within 1%, regardless of the length of the path or placement of the initial conditions within the above limits. However, from a flyability aspect the Rutowski path contains basic inherent discrepancies because connecting the points of tangency with constant energy segments produces violent gradients in control when attempting to follow the path. A pilot, when attempting to follow such a path, might think he was doing it and even state that this was done when in actuality his smoothing and alterations of the profile would shift it close to the paths marked EM-SDF. References 3 and 6 suggest a "rule of thumb" technique to modify the constant energy segments of the path result is that the EM path is 14.2 seconds or 8.2% optimistic in time and 14

and improve flyahility. The rueI is as fol.loWs:

Along constant energy segments the specific energy level is to be increased according to the altitude lost or gained in the maneuver by: AEs where K KAh -1 ifMach number is increasing (dive) (22)

K = 2/3 if Mach number is decreasing (climb).

17

0~0

*1.

0I

N I. I*0

Ca4 /

00

L
00 0

/
-

M-

/ nu

This approach was applied to the initial and final dives of Maneuver 3 where the biggest difference in results was noted previously. path is noted by the hatched line shown in Figure 5. changed path is indeed closer to the SDF path. The modified portion of the The figure indicates the

It is interesting to note that the

SDF path undershoots or, In other words, converts potential to kinetic energy faster than the E-M modified path once an appreciable negative flight path angle has been generated. From Equations (9) and (10), in summation form, the net change in the time required for the path can be written as: Et Ps - PS$*

fs7

) AE

(23)

and the change in fuel by:

Aw

Es

(Ps/*f)

-(P.i

where the asterisk indicates the EM modified values. Using an energy level incrcner.t of 2000 feet the time calculation showed a 10. 2 second incre'ise in the initial dive and a 0. 7 second increase in the terminal dive for a net At = 10.9 seconds. for a net change of Awf = - 57 lbs. The fuel calculation showed a 113 pound decrease in the initial dive and a 56 pound increase in the terminal dive Totals then are 169. 1 seconds for time The and 2740 lbs for fuel, these values are shown as "EM - MOD" in Table I. time result closer to the SDF path, however, the fuel correction creates a bigger differential. This correction was not applied to the rest of the maneuvers If the as the basic program provided good agreement between final results.

modification of the maneuver does make the path more flysble and provides a

constant energy segments were allowed to increase even more than in Maneuver 3 some type of correcting factor would definitely be required to make the EM more flyable. b. Optimality

To investigate how closely the EM paths approach the optimum minimum time schedule the three maneuvers were run on AFFDL's Trajectory Optimization Program (TOP). The TOP program provides integrated constrained 19

%j

Cd

(1)

*Q

OD N

<0

a-

00

20

C6N

A-t0
*

I1

N7x
-0 -

optinlum solutions to the maneuvers under consideration. in Figures 6 through 8 and are labelcd "TOP" in T.-hic I,

The results are shown

Figure 6 shows the comparison between the TOP and EM paths for Maneuver 1. The major elements and differences in the shape of the paths are: (1) both paths show a near level acceleration for 0.25 < M1< 0.9 with the TOP doing a gentle pullup, (2) the TOP path performs a supersonic climb crossing Mach = 1 at about 6000 feet altitude while the EM path performs a constant Mach number subsonic climb at M = 0. 95, (3) the TOP path does a pushover and almost level acceleration for 1. 1 < M < 1.4 at about 27, 500 feet altitude while the EM schedule indicates a dive with pullout at about 17, 000 feet altitude and M = 1.18 followed by a climb, (4) from 1.4 < M < 2 the paths are very similar with both showing the hook in the very end due to the poor level acceleration capability of the F-4C at high speed and altitude. Although tie paths differ considerably in the transonic and low supersonic range the agreement in final results was good with the E M method being 3. 3 seconds or about 1% optimistic in time and 156 pounds or 3% pessimistic in fuel consumption. these results are in such gPood apreement, although le paths The fact that
diffe.., i"-di.ates

a fairly wide region in the transonic - low supersonic area where there is low sensitivity of the payoff (time) to changes in the flight profile. The EM path is generally lower in altitude than the TOP accounting for its higher fuel consumption pr2diction although shorter time. Figure 7 shows a comparison between the TOP and EM for Maneuver 2. The major elements in the shape of the path are: (I) the EM path has an initial div, ig constant energy segment to the subsonic Rutowski path at M = 0.95 followed by another diving constant energy transition to the supersonic path with pullout at about 17, 000 feet and M = 1. 18. The TOP has a continuous dive with pullout occuring at about 20, 000 feet and Mach = 1. 35, (2) for Mach numbers greater than 1.5 both paths are about the same as in Maneuver 1 with both exhibiting the hook in the end for 1.9 < M < 2. 6 pounds or 0. 2% pessimistic in fuel. Agreement in results is again quite good with the EM method being 2.7 seconds or 1% optimistic in time and

21

IIS:

'NN

--

0)

hFl

cr

Cd

r0 I2

("o)

222

CTI

.W

LL LL

01

.0

AI 0

CL ,
4

cX

400

23

a,,

/j

L)

.0

/ui

II JN

24(

Figure 8 shows the two paths for mn-aneuver 3.

For the EM schedule the

initial constant energy transition bypasses completely the subsonit portion of the Rutowski path (being at a higher cuergy level) anid intersects the supersonic portion at about M = 1. 22 and 18, 000 feet while the TOP dives and pulls out at For Mach numbers greater than 1.5 the paths

about 23, 000 feet and M= 1.4.

arc close together and both exhibit a small hook at the end rather than a constant altitude acceleration. Agreement in final results is not as good with tile

EM being 10.7 seconds or (i6,%optimistic in time and 113 pounds or 4% pessiInistic in fuel.

c.

Discussion

Since it is tznown that for a minimun time problem where kinetic and potential energy are to be exchanged at no net change in energv ("pure dive or zoom) tile basic EM technique would indicate 0 seconds or 100% error, an additional problem was formulated to provide a data point between the above maneuvers and a pure energy exchange. The problem chosen was to obtain a minimum

time path bltAven an initial state of

Figiure 9 shows the two paths.

ii = 35, 000 feet,

.6,

0
=free.

W = 38, 400 lbs and a terminal state of hi = 44, 000 feet, M = 1.2, y

The EM path consists of a portion of the Rlutowski

path connected by constant energy segments to the initial and final conditions. The Tel' path dives through the EM path pulling up at about M = 1.4 and i - 20, 000 feet. seconds. The results for this problem are; EM-79 seconds, TOP-104

Thus, the EM path is 25 seconds or 24% optimistic.

It is interesting

to note that although the basic EM path is ,lot close to the TOP ill the 11 - M planec the form of the "-'rle of thumb" correction discussed previously would warp the shape of the path in the proper direction.

The percentage difference between the EM and SDF, EM and TOP final times in all four maneuvers is plotted as a fumction of the percentage of the maneuver involving constant energy segments in Figure 10. This constant

energy percentage was calculated by the length* of the constant energy segments

* Scaled directly from Figures 2, 3, 4 and 9.

25

4P

Of~

00

U./

LL) -4,4

it fn

1,~

ap$!I

26_

30

Man.
Mn

S20

/
-

tMan.

Mon. 2

3
P/

4o

I
$

////
-pilt

P(EM

lop S1

002

040

50

60

70

Constont

Energy

PortiOn

Fig-uze 10.

Pcrc~cnt Deviation iLi ]M Results-Category I,

Mtinimum

Time Maneuvers

27

divided by the total path length in tle h - M plane.

Figure 10 shows that as 0he The curve is rela-

percentage of the path invoxving constant energy exchanges increases, the accuracy of the time predicted by the EM method decreases. tively flat and accuracy good where constant energy exchanges arc less than

"r35(. Thi is I followed bL a rather steeply rising curve for the higher percentag"Os which, in the limit, approaches 100% result deviation as the constant enrlgy percenltag approaches 100%(. The curve indicates that for constant energy percentages above .10 to 50"/6 some type of correction or modification to the path is necessary to maintain a reasonable level of accuracy or optimality in the final results. 2. MINIMUTM IFUEL PATHS a. Throttle Switching are defined by:

Contours of maneuvering energy (ErI)

F II

PT s/<vf 8

(25)1

'1 Ih' contoours (can be calculated in a similar manner as the P min ininin

contours.

The

fuel parth eoan be computed in an similar manner as the minimum time Contour plots of Em and the resultant Hutowski As the objective of the EM are shown in Figures 11, 12, and 13 for maximum A/13,

iiath discussed previously.

1 pat-h>s

for tLci, F-4

,/

minllMuM

and military power respectively.

n1iiiuniu fuel path is to minimize the integral of Equation (10), this is equivalent to minimizing dw/dE or %'f/Ps along the path. Figure 14 shows the resuits of plotting xsy/t 5 versus energy level along 1 g Rutowski minimum fuel piaths for the three throttle settings. The energy levels where it is beneficial
to change the throttle setting, as shown in the figure, are obtained by noting

the crossing points of the curves to minimize the net area under the curve. 'Fijs approach could tc extended to more discrete throttle settings or a continuous throttle variation. a.love. However, this study considered only the three listed The curve shows a smooth increase in throttle setting with increasing

specitic energy, level changing from military power to minimum A/B at ES :12, (000 feet and from minimum A/B to maximum A/B at E = 52, 000 feet. 28

tI

Pi

1 4

All,

T"I

VH I'

bD

j, INN, TS ::ii t

LL. 77 77-,

or

...........

29

-77 -77T Eli T IT' 7 1: ...

77 -M

777

ilF

7
'77i fif I V. j 141

44

, T I

1 If, fli, t

77

:771

T-T

ELFit +
7

R fill

I t
.7777

ZZ

1H

F-4 4 11 4" 1

IT KA

AU Jill

Tr :m

!7 7, 'V.., Nii

+;.:! T:t -

it I., TTI I ff.l 1, rl Ll

'it :j;l :;:i

I T,P"
o: 7 L-

-,it
Pigun, 12, M alle Ilve.ring 1-:11'ergy (F

Tii T

-UTT

F." Ell;Tl

Contour Map Minimum A/11 110-wer

30

If t ..~:~

1fV1

lThttt

Fi gure 13.

Maneutve ring, F~ncrgy (E n) Contour Map I-lilitaiy Ipomer

31

7-;~

-7

iFV A SRTFAlA

4 ZVI .11 5*cll ENEG

I'*ot( ce Mii u .*fiiu .l:n ti

IlP

%4~)Klr

:>j.72

If the contours shown in Figures 11, 12, and 13 are overlayed on each other, contours of relative advantage, or best Ps/V, another at a 1 g load factor mnay be generated. of relative advantage obtained by the procedure. of one throttle setting over Figure 15 indicates the regions For example, in the region

marked MIL POWER the P / Nf values are higher in military power than they would be in either of the other two throttle settings. This type of plot would The dotted path shown

indicate to a pilot the regions in which the various power settings are most efficient in terms of fuel required for increasing energy. in the figure is the Rutowski path for maximum A/B power and the dashed line represents the three throttle minimum fuel path obtained as described in the previous paragraph. The figure shows the three throttle path contains constant energy transitions at the two throttle switching points. In the SDF simulation of the EM paths the

dives were started slightly before the switching energy level and the throttle was changed at the boundaries of the advantage regions of Figure 15 rather than at the energy levels of Figure 14. For the basic EM computation these constant

energy +transitionsat, the swicLhng points take place in zero time so the throttle increment points were at the energy level as indicated by the dashed path. b. Maneuver Results

Figures 16, 17, and 18 show the paths for Maneuvers 1, 2, and 3 and the end results are tabulated in Table I. In Maneuver 1, Figure 16, the SDF and For comparison EM paths are close together with the biggest differences oceuring at the throttle

switching pond.s and. i

tlhe

transonile consant energy segment.

the maximum A/B path is also shown on the figure.

The EM max A/B path

indicates 4687 pounds of fuel required while the three throttle path indicated 4294 lbs, a saving of 393 pounds or over 8% for the varying throttle analysis. The total time required increased considerably from 366 seconds for the max A/B to 538 seconds for the three throttle method. Comparing the SDF to the EM three throttle path the EM fuel prediction was 284 lbs or 6.2% optimistic, the time required for the paths was 538 and 539 seconds respectively.

33

['4C AlCACrr
-

V%3a,4'0 LBS
V~IEL PAiT 1

3 T tiROTrri.
MAA A18

AliIV AIA'

F'tJ:L PATH

so A-A

1ANSI1T)OAI /.FJ Ale TO

LEVEL EN/ERGY MIAX A/a

MAX

A/8

%,30

MILI

MACIVA\
Pigl-! 5.Boios or es FwlF~fii(ncN-:

I
Trotl Aalsi

34

9;4

7=.

777

. ; :... . - ... .... so . ...

77-7 ... ........

a.

77.

.. ... .....

14 lit;F
91,

ti ..........

7 I T-I

.... ..
ii:: ij Iz. Ltd I . ... ...

F5

w ai T :tji i, A

..... ... ....

TT!
771: 7.I q 77,j

7..

1, 'M

-7 T
.. 77, 7.-1 ... 7-1 :77 77 + -7 -4"77

....... at w
7 -77 7:

771

.. ... ..
"7; -1: :7:

.1777T

-77' -77-

-77-

.. ........

7:7

77,. ... ..

:r ...... . .... .7 .. . ... . ps; 7 77 7a at o .. . 7' T ..... .. ........ 1 L:

77 "'.T

k, 7:, 7,:-: 7:7. 7. 7 ZL T-7

--

77

-1

'..........

r~~-t

-Y,

.. ... ...
i~7:7

~i

7
... . .. .. .. .

7
.... ...... .. . ... . .. ...

.... .....
.. . ... ... 7 .. 7-. -7

. ........
. .; .... ... .77 :7-.: '7 -. .. . .... :

7,7 -7 T: -T:-.

71-ii i;::
7T 7... .......

..... .... ...

-77-

j.

_0

... -'.t7 777 ........

... ..... .. 777 ... ........

-T 7-:TT-

tv

..... ... . . . ... 7 :-7 .. .. .. .. ..

cl .... ... .... .. ........ .... ...

4,

7-

'IT.

... ... ....

's ... .:! :::: :::: ::!! I : 77, .... ra.;

-]A. T .. ....

:.. :::; .,! iiii

... ... ... .. M:

... ....... 4

...

... .... .. .. .... I

!-;t -7T-

4 i+ 4 .... .... .... ::I:

...........

7-7-T -'T 7- r"

-77. :U: :::I

114.14

.7-777 cl

-;LL
a 7 am

.. ... .....
7

-77

-7

._

. -

-;-=

. .. . . .. . .. . . . .
... . '7

.. .

.....

777 --7 ... ... . ... ... ...

m7r

=
. .. .. .. .. ... ... '77777 :77. . .. .. .. ... .

7. ....... .... .

T7 p:

77
77'

-3
7

... .....

37

I ho Elld and 51W Ala ncuvxe

in it ial co1rIitofls WUi

2 pahins are showii n

Fi gure 1 7,

Since tHie

hi an a rea of F`igure 15 wler ye11dL power is more efficicnt

the SM-) ,-inmlam1:ion emplovcd all thMe thrttle setting~s as indicated onl the figuirc
\\ hilc tiw I.-A i.ed p1 th .4 a~t: (,ood(I a get

Mny inh and mnax AM

A1~gain agreemlent in the shape of tile

munv

xeep]t ill flt, thromtile switclhip, and transonic areas. Good "as obtained in the fNe required wit the hA1 being 93 pounds or T1hc timne for the( S111 path was 63seconds longer than the m nHry lit power segment. inl Figulre 18 with, good path agreemient As in Mlaneuvor 2 the S1)F path, bincldes As inl the

2!.W, opiti mstic.

1-11M dwe p rohably to the iiitia

Th10 Mane11uver 3 p~aths arcV ShOwnN exetin the constant energy punjtions

A 11h1wi' in itial military power segment wiiilo the EFM path does not.
Or thiw M(0 iC)lL maevr OVC'ith

nun iiimiani timec ease this inaneux'er 6howed the poorest correlation of results of thle I Al being 179
1 )oUndb

or 7. 9%j optimiistic.

A;,:uin the Si 1ypti "us longer, taking 28 miore seconds than the EMA.
C.-

Discussion 4oat\-I i ilj Ilh PAttiO S'hoxtd a defilitk, a civanit age over utsing Only 111.1 i-

tic

111.m1

VTB puweVr for ob~taining the miiiinimum fuiel path foi, all three maneuvers I techniffue shows reasonable aectiraev where constant en r,, inIItic 'Ind poleiuti:uu comprise lesthanl 50",:as shown in Ole > I\ten

>fhi'lit 0.'U'
\W~W 5

37 47

2.63 7,9

As
I'

in t

Iionimuun time ease, it is

Inn indicates

knownvi

that in a pulre energy exehange the Therecfore, again

0 lbs fuel requtiried or a 1ioo%rvor.

LCrtiOll 10 the E Al t{Ujhniquo wmild be necessary for higher constant -0", to vc rnoaWccizcies. '0'ii '" 1 TVhe pathi ditlecreluvs he1

1SU

cv u

not as pr'loelkwcel Dintemj OwWimu

in Aw paths as

38

they were in the minimum time cases due to the higher altitudes, lower flight path angles, and lower accelerations during the climb portions of 'he pJiath.

3.

ZOOM MANEUVER The final vertical plane maneuver selected was a zoom to maximize alti-

tude.

Figure 1 indicates that the hilghest energy level that can be obtained In a pure

within the steady state operating envelope is about 107, 000 feet. would, therefore, be the forecast maximum altitude. the best pull-up angle to employ. Straight up or

energy exchange, converting all the energy to potential energy, 107, 000 feet This forecast does not 90' would be required to The starting conditions se-

account for any losses in the pull-up to initiate the zoom nor does it indicate

convert all thie kinetic energy to potential energy. (E 8, 107,000 feet).

lected for this problem were Mach= 2, 50, 000 feet altitude, and level flight

To investigate how much of the energy could be converted to altitude and

find the most efficient means of doing it, the problem was first run on the

TOP prograim.

Tihe resulting path is shown in Figure 19 and shows a rather The TOP solution is a dive with a pullout at the engine placThe maximum load factor for the Maximum altitude attained was

surprising result.

ard limits before starting the actual zoom.

path was 4 g's occuring at the pull-up altitude.

92, 000 feet leaving about 2, 000 feet of kinetic energy uncoverted. A series of inniediate pull-up zooms was then run on the SDF program to investigate the effect of a pull-up in lieu of the the initial dive generated by the TOP program. Constant, increasing, and decreasing load factors were used. Tile highest altitude attained was 88, 680 feet using a decreasing lead factor schedule with an initial value of 2.9. Finally the rule of thumb correction for energy exchanges of Equation (22) was applied to the starting conditions. This 1'orreetion, as noted earlier, proThe result of and vided a set amount of energy loss for each foot of climb altitude.

this calculation produces the path shown in Figure 19 as "EM modified" provides a forecast of 85, 000 feet for the maximum altitude.

39

L
*+

I Top Solution
f-

EM Modifie d

loo Point Con7sKTan

90
-_____

Eergy Level

80

-_

A ' i-

_______

F'niiI

"*-~~o~Enr

Number

1. Z0111 M~imie~lttudt Fi~lre lJIUVOIto

40

APcomparison of the results of the above calculations shows that a 107, 000 foot altitude (all energy converted to potential) is not achievable. Application of the rule of thumb gives too low an estimate and this rule provides a set amount of loss for each foot of climb whereas the integrated programs (TOP and SDF) indicate most of the losses are in the first 10, 000 feet of the zoom. Tie optimized path produced by the TOP program showed that, in this instance, an increase in final altitude would be achieved by an initial diving maneuver before starting the zoom.

41

SECTION IV OUT OF PLANE MANEUVERS

The initial and final conditions for the two maneuvers selected are shown in columns 2 and 3 of Table III, both maneuvers are climb accelerations with a 180' turn. Column 4 shows the method used and columns 5 and 6 the results

which will be discussed in the following paragraphs.

1.

MINIMUM TIME PATHS a. Turning Charts

Implementing the method outlined in Section 11, Equation (17) was plotted for selected values of load factor and is shown in Figure 20. Rutowski paths for the F-4C at maximum A/B power were then run for the two maneuvers at Ig and the load factors plotted on Figure 20. into Breaking the constant load factor paths down

A E/w

increm-ents of 5000 feet each, d k/dt was then read from Figure 20 Table IV shows a sample cal-

for the average velocity between energy levels.

culation for a load factor of 1. 5 -- column 1 shows the mean specific! energy level of the segments (here line I is the average betveen E/W = 5000 and 10, 000) column 2 is the average velocit ol the Jlutowski path between the initial and final energy level of the segment (lino 1 shows 670 fps) and column 3 shows the turning rate as read from Figure 20 (reading Figure 20 for liie 1 the load factor 1. 5 line intersects a velocity cf (370 fps at a ttnrning rate of 3. 05 deg/sec).

a *i is iealculated by Equation (18) as follows: the time required to go between

the energy levels along a 1. 5 g Rutow ski path is shown in columnn 4 (lun 1 shows a linme of 9.64 seconds was required to go between E/W = 5000 and E/WM = 10,000),

A %ij is

then columnn 3 multiplied by column 4 and is shown in column 5.

The

time penalty to operate at a load factor of 1.5 instead of 1.0 is calculated by Equation (19) as follows: Colunu, 6 snows the time required to go betwveen

energy levels along a 1 g Rutowski path (line 1 shows 9.40 see) and colunmn 7 the penalty time, column 4 minus column 6, which is to1 be plotted in' Figure 21 (tine one shows 9.64 - 9.40 = 0.24 see). The turning, efficiency, Equation (20),

is then eclnim 5 divided by column 7 and is shown in column 8 (line I shows

29. 4/, 24 = 1.22' turn/sec penalty).

Figure 21 shows the penalty time for the 42

TABLE CATEGORY II

III

- OUT OF PLANE MANEUVERS

Type

Initial Conditions

Final Conditions

Procedure

Time

Fuel

Maneuver I

Min Time

M - 0.5 h - 500 Ft w - 38400 LBS

M-

1.95

EM-No Turn EM-turn SDF-No Turn SDF-Turn

219 220.8 229 231 230.8 422

--

3954
--

h - 35K Ft

4t -

1800

4079 4178 4056 3216 3224 3352 3363

y=

00 00

S=

TOP-turn EM-No turn EM-turn SDF-No turn SDF-turn

Min Fuel

428 429

Maneuver 2 Min Time

I

= 0.7

M = 1.95 h = 45K Ft q/ 1800

EM-No turn EM-turn SF-No turn SDF-Turn TOP-turn

228 234.9 240.4 247.4 240.3

3720
--

h - 30K Ft W = 38400 LBS

3767 3897 3810

y=

0' 0

4=
Min Fuel

EM-No turn EM-turn SDF-No turn SDF-turn

330
--

3183 3243 3301 3351

355 359

43

-77 -77 mo; 77,- 77n --- 7r -- 7 .. ........ L" ..

-777

... ........ 7!-- - -I.. .... 7 I

IT
............

r! --Fil17

bT .

::

IT
4:1 T

-3:
'44
:7,T
7-,

- ! ....

it

TI
-.. 7 rii; 77.-

77,
U1. -77 -I'll cr.

7 T 7r. -Iii LE
:!t! MT I

:4 ..

i,it

T: 7

Cd

ji:j: HT .

-till, ... . ....

7t
7-777 T 7-. i. 77t7!7 7 t7177 .. . .. .. .... .... ..

41 -4-m
.7

;T4 j; 7

. ... ..... .... .... .... ... .... -:.`!

'T -7=

t-44 ... ..... .... 7r!t -7777 -717

r -77! .4-r, 771 ;+ 7 .. ..... 71T. I .. .. .. .. ... . ... . . tA :!, 7: 77 7...

... . .. ....
r

77. A :

-77

........

... . ..

77 77

7,P:1
P. 7 .

d. 't 7:77 T7.- ::i i7i 7 .

..........

44

14-

*00

1j~

0.0N

44 4-,,

~)4r (' ~O
Li

4.)

4L-

0:

,S

4.

co~

C4

r-

1*~

r-I

%O

~45

P ;w

i
.-

-7

I,

iF~
1

~ J-c1-qI

T'7

4 -N

46

series of load factors.

For the sample calculation the time penalty of 0. 24 For the

seconds is plotted at a load factor of 1.5 and specific energy level of 7500 feet. Figure 22 shows the turning efficiency for the series of load factors. sample calculation TE = 122 is plotted at load factor = 1.5 and E/W = 7500 feet.

Figure 22 shows, as would be expected, the highest values of turning efficiency are at low speeds and low load factors. At the very low energy levels there is a reduction in efficiency due to the low level of excess power available for acceleration causing the hump in the curves. The "constant efficiency"

turns used correspond to horizontal lines across this figure producing a table of load factor versus energy level. From this load factor table Figure 21 can be

read to obtain the penalty time and the amount of turn calculated by Equation 20. For Maneuver 1 a 1800 turn between energy states (E/W)i, = 5, 000 feet and (E/W)f = 91, 000 feet is required. To start the calculation a turning effiFigure 22 is then read at the various The turn was conFigure 21 is then The

ciency is assumed, say 1030/see penalty. sAide red complet L~kU byI an )
62,00 UL

energy levels with the results shown in Table V, column 2.

ft ab the Load

Lace dropped below

1. 01 g's which was the lowest considered in the analysis.

read to produce the penalty times for each segment shown in column 3 (line one shows 0.40 seconds penalty for a load factor of 1.73 at an E/W of 7500). and this is shown in column 4 (line 1 shows bank = cos I ( 1/1.73) = 54. 7), In general The

load factors can then be converted to bank angles through the use of Equation 15 Summing column 3 the total penalty time is 1.74 seconds and the corresponding amount of turn achieved through Equation 20 is 103 x 1.74 - 179. 2. the first guess at the turning efficiency would not be this close and a graph such as Figure 23 could be made by plotting the results of several calculations.

solid line shows the amount of turn achieved between specific energy levels of 5, 000 and 91, 000 feet for varying values of turning efficiency. shows the corresponding time p( nalty. The dashed line Our sample calculation is shown by the

47

j :7 . . ...

VMV.

117r
I

,tIC-!

Islo

<!

1
.. .'..... ...

47.

TABLE V MINIMUM TIME CONSTANT EFFICIENCY TURN EXAMPLE CATEGORY II MANEUVER 1

=

103 deg/sec penalty

A E/w increment = 5000 ft.

Mean E/W (ft) 7500 12500 17500 22500 27500 32500 37500 42500 47500 52500 57500 G2,u00 n (g's) from fig 22 1.73 2.10 1.85 1.58 1.3 1.19 1.085 1.033 1.022 1.014 1.011 1.009 Total Atp
-

t sec) from fig 21 .40 .31 .20 .19 .17 .15 .10 .07 .06 .04 .03 .02 1.74 sec

Bank(deg) Eqn 15 54.7 61.5 57.3 50.7 39.7 32.9 22.9 14.8 12.1 10.0 8.9 7.7

491

7: _ T,

IToi
-

f 4;1 2t 4I7 L.A~j

* HI
~*T
i

.1 1K]
~

.;F4

4 h{*'I
I

1i77

,14-i

1i

50

heavy lines.

The family of bank curves generated by varying values of turning

efficiency for this maneuver is shown in Figure 24. b. Flyability and Accuracy

The procedure used to investigatc the accuracy and flyability of these maneuvers was the same as the method of the previous section, with the addition of the bank angle. The SDF program was used first to simulate the no turn (zero bank) Rutowski path and then the bank angle arrived at by the EM turning procedure was inserted as a function of time and the SDF program run again using the same angle of attack control. Figure 25 shows the resultant paths for The The SDF path Maneuver 1 and the results are tabulated in columns 5 and 6 of Table III. EM path with no turn was 219 seconds, with turn 220.8 seconds. with no turn was 229 seconds and with the turn 231 seconds.

This result in-

dicates that the turning penalty calculated is a realistic value and also that the EM path underpredicted the time required for the maneuve r by 10.2 seconds or 4%. The EM path underpredicted the fuel required on the no turn path by The point where the i80W turn was completed on the SDF program, At this point the bank angle in the SDF program was set to 97 lbs or 2%'L,.

marked on Figures 24 and 25, was reached slightly before the 1.01 g load factor prediction point.

zero for the rest of the path. The calculations for Maneuver 2, which starts and ends at a higher energy level, were done in the same manner as Maneuver 1 using Figures 20, 21, and 22. The family of curves shown in Figure 26 were generated for this maneuver. As shown in the figure the efficiency index for a 180' turn was down to 26" turn/second penalty in contrast to the efficiency of 1030 turn/second used in Maneuver 1. This decrease in turning efficiency is due to the much higher Note in Figure 26 that an efficiency of For the SDF starting energy level for Maneuver 2.

100 for this maneuver produces a turn of only 40" heading change. til the turn was completed. of the path.

simulation when the minimum bank of 7.5' was reached this value was held unThe bank angle was then set to zero for the rest The path results are shown in Figure 27 and tabulation in Table HI.

The EM path with no turn was 228 seconds and with the turn correction 234.9 seconds; a penalty of 180/26 or 6.9 seconds. The SDF path for the maneuver with no turn took 240.4 seconds, with the bank schedule for the turn 51

I
_ _

OD_ )

-0
InI
90 CL 0. at rd II

LL..

LU

IT 0
Tr,_

0
-'V N_

C) 0

.6(]a lbuv iNUcg

1
0
52

0-

ci

... .. ... ...

.... ....
.......

...

it

....
if

If 7T Itl
,T if i .1

7r 7MT :s! itI

fill

fit

,T

If i iIf 11

ff. AiT ;;1 H

:it .41: t, 11

HI I

in!
1"A 4 H!

Nil

flji

fit

!H
0,; jj,Tlijj -1111-144.

if :.I

it 4 I 4! j: iil i!fl I"IH !t if tjgii

j4,

TTT If i :11: ill!

TT

'; Ai:

7t

-114

-t

it. IN! IM,

I't

if i!i*
t"r!T If -7 7. IT m a 7777 iii! i-ii i

ll

till 11, If:, :fit !i. TIT:iR i!N IT "7r,

bD Cd H

:il: if

to f:!

.. . .

C-4

i :it: i:if

:jp

I Fit
7::-,TT 7-r7 7:7-t -t.................... im.. .... .. . ... . ... .. .......... Tf-

iT !i1i

Li

.17

7T .. f, : ;! 5 7,rl 71",

.... t i i , ::. !;: ;I-il 1

15 7t=77-T

:j ! iiiil

MT

L7.

53

Constant Turning Efficiency Analysis ,E/w|i 60 F4C Maximum A/B Power I- Heading Chonge 320 -1 Dog. 1 I I (16) 50 --Efficiency Index Deg/Sec Penalty

r (W) (E

'

4C

..

___-

I
(100

"
\40 \

Minijmu z0 30 40

BanK Considered"S

50 60 70 80 Specific Energy Level (E/W) * KFT

90

100

I- I:,'V(. 26'.

Category 11, Maneuver 2, Miniinuin Time-Bank Schedule Farily

54

...

...

.. ..

. ..

. .p..

77,T

In:

p ,HF:Ijt, 1
77

t~ci

5rr4

H!U

"2A7.4

1. an , !oids,

|.urninl w'

tium- pjenalty of 7 seconds.

As in MIaneuver 1 the Again the EM

t, 'nif", penalty calculated by the use of .he charts (6.9 seconds) appears to be a r.:isonitblc value when compared to the SDF result of 7 seconds. path unidcrprcdictcd thle totl time required for the maneuver this time by 12.5 sL'coldS,, (or 5%. The fuel required on the no turn case agaill showed good agr('e mea wAth the I vT procedure, underpredicting by 48 lbs or a little more

c.

Optixnalitv

To investigate how closely the flight paths and results of tile turning protudu e qpp)roach the optimum miiiiinium time maneuvers these maneuvers were

run on the 'FOP proecam.

The results arc labeled "TOP

"

in Table I1. The dashed line

"solid 1r: Figaure 2S shows the TOP path for Maneuver 1 and tei
sh;ow,vs the EM path for comparison,
sptq)ervsoI`.(. cliin

As in Category 1, Maneuver 1, the principle

dilcrencecs in the paths in the. h - M plane are that the TOP Path performs a: while the EM is subsonic and that the TOP path does-, a pushover The EM palh with turn correction as noted preaind olinst level -ceeleratiou between 1.1 < NI < 1.4 while the EM path shows a .',.,,vCd dive a,d lpuflup. vioc sly predicted 220. 8 seconds while the constrained optimuni. as computed by fla TOP prograynm was 230.8, the EM tmiderpredicting the time required by ,..,; , 11.i. J"i''sirc 29 shows a Lank angle compaprison between the ,wo .l iho.,b liv\bh n" sehdubs arc considerayly differenit, in both cases

S;.,

il. o0 'he htirm is conqdctc-l in the first 80 seconde of the maneuver.

Tki? solid line in Figure 30 shox s the TOP path for Maneuver 2 and the ,i,:.,i,. :,,s. the E.P path for compa'riscn. A p1id wift ume "'fu IV: .ompita.d The principle differeince in the paths n-, U, Al ph.m, as in Category I Maneuver 2, was in the initial diving percorrection predil'ted 28,4.9 seconds while the TOP was 2410.3 seconds, the EM underpredicting the

*,i:,L"mn,5; 0114
l300

Figutre 31 shows tile bank schedules from .,, Iw- methods which are again considerably different. The TOl- result again
time )y 5.4 s-conos- or 2%/o.

,, o,,t of the torn completed in the first 80 seconds while the constant

56

I-I
w 2
-

I____
___
aptC1

o5

CC

04 *00

F4'

______ ____

___CC

0o
saa~isp-alb v lIf

58

_________

-00

Cd '3)

U))

C5 Fn-

_____

'.-W

CLC

4',4

NrN

60

efficiency turn uses bank angles greater than 200 for the first 140 seconds of the
4

marieuver. d. Discussion

In the two turning maneuvers studied good agreement in the final results was obtained between the EM, SDF, and TOP simulations. The same comments The on the basic nonturning paths as were made in Section I1, 1. c apply here. the SDF simulation in these maneuvers.

turning procedure appears to give realistic turning penalties when compared to The bank angle history was quite different from that obtained by the TOP program yet there was good agreement in the final results indicating the results were not extremely sensitive to the exact optimum bank schedule. 2. MINIMUM FUEL PATHS a. The Turning Charts inimunum fuel turns were treated in a similai manner to the mininum However,

tinie turns described previously u ing minimum fuel Rutowski paths.

the procedure is somnwhat longer when several throttle positions are considered. Minimum fuel ]lutowdki paths were run for military, minimum A/B, and maximum A/B, at the load factors plotted on Figure 20 up through 1. 5 g's. of Section III, 2. a. Table VI shows a sample calculation to develop the minimum fuel turning efficiency charts for a 1. 1 g load factor. Column 1 in the table shows power setting, column 2 the energy segment considered, column 3 the average velocity of the 1. 1 g Rutowski path between the initial and final energy level of the segment, column 4 is the turning rate as read from Figure 20. the energy levels of the segment along a 1. 1 g Rutowski path unn 4 multiplied by coluni 5 and is shown in column 6. The throttle switching points between power settings were determined by the method

A *$is then calA qi is then col-

culated by Equation (18) as follows: column 5 is the time required to go between The fuel penalty to op-

erate at a load factor of 1.1 instead of 1.0 is calculated by Equation 19 as follows: column 7 shows the fuel required for the Rutowskl path segment at a

61

"1 C4

(' 44 N

0 -1 r,

C rN co .-

C3 4 -4

co O

co C

L) (

C.4 -

17

(N r4

41

C14

CO q

(4

C11 ,4

0D H 0 (r0-4 C4 ' co C4

In '-4 4 C-1

(N

10

C
4(%

It N(

0 CO

a, IT

.4 0" .1s ,4 04 LM4

(N4 OD N %0 CO '4 %0 (N H4

4'4 4(4 14 (N CI Os 64~~1 C " Hl 0 rH

C" C 00 14 -4

14

0 ' co rH

0 0 m~ H-

H0 -. OH

(n

00 a H

'0)
41

~4

ol

'(4 e.1 0

In

CO 4

co

0 vi

H m(

H 4

so
0 0

Nn -

Hr-4

rC "

C"

O.
0p

A0
C"

'0 N' IN C4

(N "

10
(

a)

N (N C"

C"cl r CO.

(Nl

'0
cc. H

'

'0

H 0 QO (N

a
1"

as

as

4-4

C'

(3)

(('C

W)
. .

0
.

0I
.

0:
.

(n

C"

m-~C

L..

as

as <1

:lr

-4

-4

load factor of 1. 1, column 8 the fuel required at a load factor of 1.0, and column 9 is the fuel penalty to operate at a load factor of 1. 1 instead of 1. 0 calculated by column 7 minus column 8. is then column 6 divided by column 9, The fuel turning efficiency, Equation (20), Figure 32 shows the fuel penalty time calFigure 33 shows the

culations for the series of load factors up through 1.5.

turning efficiency in degrees of turn per pound of fuel penalty. Figure 33 shows, as in the time case, that the best efficiencies are at the lower load factors and speeds. The crosshatched lines show the throttle switchAs before, the ing boundaries for best efficiency along the Rutowski paths.

"constant efficiency" turns correspond to horizontal lines across this figure, each efficiency level producing a table of load factors versus energy levels which, by using Figure 32 and Equation (20), can be resolved into the fuel penalty and amount of turn. penalty. Using Maneuver 2 as an exampie an efficiency level of 3 degree turn/pound penalty was assumed. Table VII shows the results of the calculations. Column 1 in the table shows the energy segment considered, column 2 the load factor read from Figure 33 (line 1 shows a load factor of 1.4 read from Figure 33 at a mean energy level of 40, 000 ft, turning efficiency of 3), Column 3 shows the fuel penalty for operation at the load factor of column 2 (line one shows 18 pounds penalty for the segment to operate at a load factor of 1.4 instead of 1. 0 read from Figure 32 at E/W = 40,000 feetand load factor of 1.4), (line 1 was calculated by cus- ( 1/1.4) - 44.4). culated by Equation 20 as 3 x 60.30 = 180.90. Column 4 shows the bank angle computed by Equation (15) from the load factors of column 2 Summing column 3 the total fuel penalty is 60.3 lbs and the corresponding amount of turn achieved is calTo generate a family of curves with different amounts of turn the above calculation could be performed for several efficiency levels and a chart similar to Figure 23 prepared. b. Maneuver Results In general, the more turn that is required between two energy levels the lower the efficiency level must be and the greater is the fuel

The pro edure used to evaluate the results was substantially the same as in Section III, 2.b with the SDF program used first to simulate the no turn 63

Ii~

.414L

Is
;A

7-

.. .... .... .. w a -7:: '7:

.... . ... ..

77.
m .a mm
i:: ... .. :!: Tlt

.... ...
w: r.' 41'

:W

it
:it U + ti. iip tm IW: i a flT

:'t Tilt

;7m

.... .... L2 Trr

na! ;

Z dL

... ; :....

44'Elfl

vi '

co

.. .... ....

MR.:.
-4
vM-.

.. . ....
TT
4 --i "i

bh

i4i+

.. .... tt. wi
.1. It 1

Nk

Li

1-'

7rr HN -.ii iF jjj qii 'I!

i 4 ... .. ... .... 4'0:77

T 75

1:3

co
E! Ekt-41, j 1u:

it -rq it A ji;i 7 Ir

44 :.
No :'14A .... ......

AJ

65

TAULL VII MIINIMIUM FULL CONSTANT CATEGORY EFFICIENCY TURN EX!*IPLE I1 MANEUVER 2

4/IAWWp

3 deg/lb pcnalty

0
E/IW,(KFT)

@
n(g's) from fig 33

01
6Wp (ibs) from fig 32 18 14 8.4 7.0 2.5 2.4 2.2 2.0 Bank (dog) Eqn 15 44.4 30.5 17.8 16.9 17.4 17.4 17,. 16.5

37.5-42.5 42.5-47.5 47.5-52.5 52.5-57.5 57.5-62.5 62.5-67.5 67.5-72.5 72.5-77.5

1.4 1.16 1.05 1.045 1.043 1.048 1.046 1.043

77.5-82.5 82. 5 S7..1.015

1.035

1.6 1.2

14.9 9.9

: 592.5

1 .0

7.6

TLaI Ail't

60.3 liiu

66

Butowski minimum fuel path then again using the bank schedule from the turning
procedure. Figure 34 shows tile resultant paths for Maneuver 1 and the results The EM no turn path required 3216 pounds of fuel are tabulated in Table III.

and with the turn 3224 pounds a difference of only 8 pounds of fuel required to do a 180' turn in conjunction with the increase in energy state. The SDF no

turn path required 3352 pounds of fuel and with the turn 3363 pounds a difference of 11 pounds required for the turn. Comparing the end result the EM procedure Figure 35 shows the bank

was 139 pounds or 4% optimistic for the maneuver. angle schedule used for Maneuvers 1 and 2.

Maneuver 1 uses steadily decreas-

ing and relatively low bank angles throughout the maneuver completing the turn before reaching supersonic speeds. Maneuver 2 starts at a higher bank and is

turning throughout most of the maneuver, the bank angle calculation was performed in the last section and is also shown in Table VII. Figure 36 shows the The SDF

results for Maneuver 2 and again the paths are quite close together.

path contains a short military power segment as noted on the figure as the initial point was off the Butowski path falling in the military power advantage region of Figure 15. The EM path starts in min A/B power since the analysis permits

an instantaneous constant energy transition to the min A/B Rutowski minimum fuel path. For this maneuver the EM no turn oath indicated
31 F,3

pnunds of fuel

and with the turn 3243 pounds a difference of 60 pounds of fuel needed for the 180 turn. The SDF simulation showed that 3301 pounds of fuel was required in

the no turn case and 3351 for the turn, a difference of 50 pounds required for the 1800 turn. Comparing the end results the EM procedure was 108 pounds or

31o optimistic for Maneuver 2.

As a further cheek on the utility of the varying throttle technique the EM analysis; of the. power.
aneaIUuvers

witLhOut the turns were also run at maximum A/B

The results for Maneuver I were 3514 pounds fuel and 264 seconds

time representing a savings of 290 pounds fuel or 8% for use of the varying throttle technique instead of maximum A/B power only. For Maneuver 2 the

result was 3235 pounds fuel and 277 seconds time showing a savings of only 52 pounds or less than 2% due to the relatively short time a power setting oti er than maxinmum A/B was used in the varying throttle analysis. The turning

analysis was not performed for the maximum power only paths, however, it would be expected considerably more fuel would be required for the turns due to the shorter length of time tile vehicles would operate in the most efficient turning area of the charts. 67

r--7

-- -.. . .

I
r...

... .... .
f N

71.17
t:i

*a

77r.

71-7 A
A-b

Ijjt~~y

... ...

L.7
: -. . .. . . .. . . .

.. .. ... .. .

0 1c~~~
4P
____

c0
U4-

C~CL

CC 4OD

oV 0

?9) Ca4

I
I-

NrN
4-

o, a.0
a Ln

00c
Iq) 0

rro
-

04-0

70

SECTION V CONC LUSIONS

The cornparison of the energy based maneuvers to the TOP and SDF solutions showed the energy paths are always optimistic in the results which can be achieved. This result optimism increased as the percentage of the path involving constant energy dives and zooms in the altitude-Mach plane was increased. For the minimum time paths the result optimism was less than 2% for the maneuvers where the constant energy percentage was less than 35% followed by a rather steeply rising curve approaching, in the limit, 1.00% error for paths which are comprised entirely of a constant energy transition. It was observed during the course of the stud,. -hat the paths run on the TOP program, with the exception of the zoom, tended to dive towards the Rutowski path, follow the path fairly closely, arid then perform a zoom or a dive to reach the prescribed end conditions. This trend was followed even when the accuracy of the EM method, as measured by the final result, had deteriorated to 24% error. it follows, therefore, that knowledge of the basic climb path, whether this path is run using EM methods or by an integrated program, can provide a pilot or analyst the approximate path and some indication of the results for a large number of maneuvers. runs. To obtain as much information through parametrics or by separate optimization for each maneuver would require a large number of It should he noted that when computing a Rutowski path. speed and altitude should be limited only by physical limits such as engine placard or buffet boundaries leaving the technique free to produce constant energy dives and zooms. An artificial limit say on the altitude not to exceed the altitude of the desired end condition would force a level acceleration at t.his altitude if the true lRutowski path was above it, producing a result which could be far from the optimum, program showed that all paths Use of a rule of thumb type However, there is no gaarantee

The simulation of tie energy paths on the SD1 required some modification t.) make them flyable. correction increased the flyability of the paths.

that the results from use of the correction will be any closer to the optimum than the unmodified path particularly on dynamic paths involving large flight path angles. This was evidenced by the zoom maneuver in which application of the 7I1

rule (A thumb changed the result from optimistic to conservative. taut for the wore dynamic mazeuvers.

The TOP

proggrtin provides flyable optiniun solutions which become increasingly impor-

The use of a variable throttl.e setting for the minimum fuel maneuvers showed a clear advantage to use of partial power settings in the regions where they are most effective rather than maximum power only. The results for the F-4C showed that in an energy climb starting at a low energy level this savings was about 393 pounds. This corresponds to an 8% ..avings in a long climb to near maxinum energy or a savings of over 20% if the path is terminated before reaching supersowe speeds. The turning analysis showed in the maneuvers investigated that turns may bc made in conjunction with a Rutowski path without suffering a severe penalty in accuracy. Tue technique has the advantage that once the turning charts are The combination of the varying throttle and developed for an aircraft, penalties and bank schedules can be rapidly calculated for a large number of maneuvers. the turning analysis provides the analyst quick insight into difficult two or three control variable problems involving pitch angle, bank angle, and throttle setting. The resultant solution can be used as a first estimate and as a nominal for a iote accurate integrated program to shorten the running time for the optiniiZ$UtiO1

process.

Tlhe findings of this study are that energy methods offer a tool especially usefuI in the early stages of preliminary design and functional performance s.udt-i where ratfpid solutions are needed and reasonable accuracy is acceptable. if the analyst uses good jucigement in applying the methods to maneuvers the results provide a good qualitative insight for comparative purposes. The path, ,.,ever, should not be used as a source of maneuver design or flight schedule in its entirety without verification, especially on relatively dynamic maneuvers where the optimnality and accuracy of the energy maneuverability procedure decreases. The determination of what method or combination of methods is best suited for a particular application involves trading off the high speed and limited accuracy of the EM teclhiques against the longer running but more :,c-uratc progams such as TOP.

72

"- -- ..

...

. ,-- "

...

'

'"

J "-

Il

REFERENCES

1.

Air Forcc Flight Dynamics Laboratory; Threc-Degrec-of.-Freedom OptIimization Program, FDL-TDR-64-1, Vol 3 Part 1 and 2, Wright Patterson AFB, Ohio, October 1964. Butowski, E. S.; LynergyApproach to the General Aircraft Performance Problem. Douglas Report SM - 14875, July 1953. Chase, Rankin, and Weatherford, Energy Maneuverability.Program Documentation Listing, Eglin AFB Fla. Jan 1968.

2. 3,

4.

Air Force Flight Dynamics Laboratory; Six-D;2gree-of-Freedom Flight Pafth Study Generalized Computer Program. FDL-TDR-64-1, Vol I Part 1 and 2, Wrigh--Patterson "F" Ohio, Octobei 1964. SESSV, Energy- Maneuverab ility Program Programmers Handbook, Wright-Patterson AFB, Ohio 1966. Dunlap, D.F., Ursel L.V., Computer-DisplayFeasibility for.Flight Performance Optimization, AFFDL-TR--68-125, Air Force Flight Dynamics Laboratory Wright Patterson AFB, Ohio, October 1968.

5, 6.

73