Добавил:
Upload Опубликованный материал нарушает ваши авторские права? Сообщите нам.
Вуз: Предмет: Файл:

Thesis - Beaver simulation

.pdf
Скачиваний:
38
Добавлен:
19.02.2016
Размер:
5.22 Mб
Скачать

Appendix F

217

Type help fmdims for on-line help a t the MATLABcommand line. The internal structure of FMdims is shown in figure F-11.

Inputvectors:

Y D ~ Iand C,

Outputvector:

F a

It is possible to use FMdims for other types of forces and moments too. For instance, this block is also used within the subsystem Engine Group, where in stead of C, and F,, C, and I?, are used.

The variables cbar, b, and S are defined in the Mask-definitions line:

cbar = GMl(1); b = GMl(2); S = GMl(3);

See table F-3 a t the end of this appendix for the definition of GM1.

C. Engine Group.

The following'basic' blocks are available in the subsystem block Engine Group, which is depicted in figure F-12:

1- Power (Beaver). This block is used to calculate the engine power P and the dimensionless increase of pressure across the propeller for the DHC-2 'Beaver'. Type help power a t the MATLABcommand line for on-line help. The internal structure of Power (Beaver) is shown in figure F-13.

Inputvectors:

x,ut and Yatm

Outputvector:

Ypow

2 - Engmod (Beaver). This block calculates the force and moment coefficients due to the operation of the powerplant (including slipstream effects), according to the model from ref.[30], see equations (2-10) in section 2.2.2. Type help engmod a t the MATLABcommand line for on-line help. The structure of Engmod (Beaver) is shown in figure F-14.

Inputvectors:

and Ypow

Outputvector:

Ct

The block Matrix Product, which has been stored in the library TOOLS, is used to multiply the help vector ytmp,which is defined as:

Ytmp - [ dpt dpt3 aodpt2a2.dpt ]

with the matrix EM. This matrix contains the coefficients of the engine forces and moments model, see the definition in table F-2 a t the end of this appendix.

3 - FMdims. This block computes the actual forces and moments, using the dimensionless coefficients and the dynamic pressure. I t already has been described a t the 'Aerodynamics Group'. See figure F-11.

d. Gravity forces.

The forces due to gravity are calculated with the block from figure F-15:

Gravit.~Assuming. that the mass of the aircraft remains constant under the motions considered, the body axes gravity forces are calculated. If the mass is not constant, it is necessary to implement the mass as a n additional input to this block. Type help gravity for on-line help a t the MATLAB command line.

Inputvectors:

and Ya,

Outputvector:

Fgr

The variable m has been defined a t the Mask-definitions line:

See table F-3 for the definition of GM1.

e. Wind forces.

It is possible to use the nonlinear equations of motion in non-steady atmosphere, if additional forces due to the wind and turbulence are included. This is done with the following block (see figure F-16):

Fwind. This block includes contributions to the body axes forces due to non-steady atmosphere. Type help fwind a t the MATLABcommand line for on-line help.

Inputvectors:

x and uw

Outputvector:

Fw

The variable m has been defined a t the Mask-definitions line:

See table F-3 for the definition of GM1.

f. Total forces and moments.

The following block is used to create two vectors, with the total body axes forces and moments respectively:

FMsort. This block first separates the moments from the forces, and then adds all forces and moments to eachother. This block only functions properly if the aerodynamic forces and moments are expressed in their body axes components. If the aerodynamic lift, drag, and sideforce are used in stead, this block needs to be changed. It is also necessary to change this block if more contributions to the total forces and moments are included.

Appendix F

219

Xgr

.-.-..

 

m*u[l7]'cos(u[8])*sin(u[9])

Mux

 

 

Ygr

 

yatm

I

m * ~

1

 

 

 

 

Figure F-15. Internal structure of the 'basic' block Gravity.

Main \ Beaver dynamics \ Gravity

J L

-Xw/m

x

Mux

 

-Yw/m

I It.#

 

Figure F-16. Internal structure of the 'basic' block Fzuind.

Main \ Beaoer dynamics \ Fwind

Mtot

Figure F-17. Internal structure of the 'basic' block FMsort.

Main \ Beaver dynamics \ FMaort

Type help fmsort a t the MATLABcommand line for on-line help. The internal structure of FMsort is shown in figure F-17.

Inputvectors:

Fa,Ft ,F, ,and F,

Outputvectors:

Ftd and M,

p. Equations of motion,

The internal structure of the block called Equations of motion is shown in figure F-18. This block contains the subsystem State derivatives, which is shown in figure F-19.

First the twelve time derivatives of the state variables are computed. In order to determine the position of the aircraft relatively to the earth, the body axes velocities of the aircraft relatively to the surrounding atmosphere are calculated and added to the wind vecolity components in the body-axes, u, ,v, , and w, . Then corrections to make the implicit equation for P explicit are made, and some state derivatives are artificially set to zero if that is required by the user. Finally, the state derivatives are integrated with respect to time, which yields the state trajectories.

The integrator block is a standard SIMULINKblock, which uses the variable xinco as initial value of the state vector (xinco = x,). The Gain block multiplies the resulting vector x element-by-element with the vector xfix of length 12, which has elements that equal either one or zero. The gain vector xfix can be changed manually, or by calling the helpfunction FIXSTATE (section F.2.6), which makes it possible to fix one or more states, i.e., set the time derivatives of one or more states to zero. This may, for instance, be convenient if the user wants to neglect longitudinal or lateral motions. By default, xfix is set to 1, which is equivalent to xfiz = [ 1 1 1 1 1 1 1 1 1 1 1 1 ] T.

The equations for the time derivatives of V, a,and P are valid only if the aerodynamic forces and moments are expressed in terms of contributions along the body axes. If the aerodynamic lift, drag, and sideforce are used in stead, it is necessary to make some minor changes to these equations, see appendix A.

The subsystem Equations of motion consists of the following blocks:

1- State derivatives. This is a subsystem block, which contains a number of 'basic' blocks, needed to compute the time derivatives of the state variables, see figure F-19. The following 'basic' blocks are included:

1.1 - Hlpfcn. This block is used to compute some sines and cosines which are used more than once in the calculation of the different state derivatives. See figure F-20 for the internal structure of this block. Type help hlpfcn a t the MATLABcommand line for on-line help.

Inputvector: x

Outputvector: Y h l p

1.2 - Vabdot. This block calculates the time derivatives of V, a , and P. On-line help is available (type help vabdot a t the MATLAB com-

Appendix F

221

 

 

+

+

 

 

 

 

Gain to fix

Integrator

State derivatives

11sxdot

states

uwind

 

yatm

ybvel

Figure F-18. Internal structure of the subsystem block Equations of motion.

Main \ Beaver dynamics \ Equations of motion

 

 

 

V dot

 

 

 

 

alpha dot

 

 

 

 

beta dot

I

 

 

 

Vabdot

 

 

I

rdot I

 

Mtot

hlpfcn

 

pqrdot

 

 

 

-

 

 

 

 

psi dot

 

 

 

+ theta dot

 

 

 

 

phi dot

 

 

 

 

eulerdot

 

 

 

-.I-s ybvel

 

 

velocities

C111

 

 

 

 

 

 

u w

 

 

DeMux

 

xe dot

 

 

 

ye dot

,-

 

>

 

 

I

H dot

I

uwind

Time derivatives

 

 

 

of windspeeds excluded...

Figure F-19. Internal structure of the subsystem block State derivatives.

Main \ Beaver dynamics \ Equations of motion \ St& derivatives

I

cos alpha

 

I

sin(u[2])

sin alpha

 

I

cos beta

 

J

sin(u[3]) ..

I

sin beta

 

I

tan beta

Mux

I

I

sin psi

cos psi

 

k

1

sin(u[8])

 

I

sin theta

 

I

cos theta

1

sin(u[9])

I

sin phi

 

- - --

 

cos phi

Figure F-20. Internal structure of the 'basic' block Hlpfcn.

Main I Beaver dynamics \ Equations of motion \ State denuotiues 1 Hlpfcn

Vdot

 

M w

alpha dot

P/dot;

 

(-u[13]*u[19]*u[22]+u[14]*u[21]-u[15]S[20]*u[22])/(m*u[1I) + u[qS[20] - u[6]*u[19]

alphadot;

betadot]

 

beta dot

 

Figure F-21. Internal structure of the 'basic' block Vabdot .

Main \ Beaver dynamics Equations of motion \ State &riuatives 1 Vnbdot

DeMux

Matrix ~ a i n

ypqr = [pdot;

Here: ypqr = GM2*ytmp

qdot; rdot]

Mtot';yhlp']'

ytmp = [L M N pA2 pq pr qA2 qr rA2]'

Figure F-22.Internal structure of the 'basic' block pqrdot.

Main \ Beaver dynamics \ Equations of motion \ St& derivatives \ pqrdot

 

(u[S]*u[28] + u[q*u[29])1~[27]

 

 

psi dot

 

 

~[5]*~[29- ~] [ 6 ] * ~ [ 2 8 ]

 

[x';

theta dot

 

 

 

Ftot';

 

thetadot;

Mtot';

u[4] + u[30]*u[26]

phidot]

Yhip]

I

4

phi dot

Figure F-23. Internal structure of the 'basic' block Eulerdot.

Main 1 Beaver dynwnics I Equations of motion I State derivatives 1 Eulerdot

u[l]'u[l9]'u[21]

 

U

 

4

 

v

Mux

 

u[l]*u[22]

 

 

[x'Ftot' Mtot'yhlp']'

 

 

ybvel = [u v w]'

Figure F-24. Internal structure of the 'basic' block uvw.

Main \

Beauer dynamics \ Equations of motion 1 Stah? derivatives \

uvw

 

u[3O]'u[27]+ (u[31]'~[28]+~[32]'~[29])'~[26]

 

[x'

 

 

 

Ftot'

 

 

 

Mtot'

tmp6

 

 

Yhlp'l'

 

 

 

ybvel

 

 

 

 

~[30]'~[26- (~[31]'~[28]+~[32]'~[29])*~[2~

 

H dot

 

yedot;

 

 

 

Hdot]

Figure F-25.Internal structure of the %asidblock zyHdot.

Main I Beauer dynamics I Equations of motion I State derivatives ( xyHdot

-

1

DeMux

Mux

xdot uncorrected

I

I

1

X

yatm

4 Mux

f(u) t----

I

beta dot

correction

Figure F-26. Internal structure of the 'basic' block xdotcorr.

Main \ Beaver dynamics \ Equations of motion \ xdotcorr

Appendix F

225

mand line). The structure of Vabdot is shown in figure F-21.

Inputvector:

[ xT FtolT MtotT yhlpT lT

Outputvector:

Yvab

The variable m has been defined a t the Mask-definitions line:

See table F-3 for the definition of GM1.

1.3 - pqrdot. This block calculates the time derivatives of the rotational velocities along the body axes, p, p, and r. 'I'ype help pqrdot a t the MATLABcommand line for on-line help. Figure F-22 shows the internal structure of pqrdot.

Inputvector:

[ xT qotT W o t T Y$ lT

Outputvector:

Y P Q ~

A Matrix Product block from the library TOOLS, which has been included on the floppy disk accompagnying this report, is used to multiply the helpvector ytmpwith the matrix GM2. This helpvector is defined as:

See table F-3 a t the end of this appendix for the definition of the matrix GM2 with inertia parameters. Notice that the mass distribution of the aircraft is considered to be constant during the motions of interest. If the mass distribution does change during these motions, it is necessary to write out the matrix multiplication, using the mass-distribution parameters as additional variables.

1.4 - Eulerdot. This block is used to calculate the Euler angles v, 0, and 9.Type help eulerdot a t the MATLABcommand line for on-line help. The internal structure of Eulerdot is depicted in figure F-23.

Inputvector:

[ xT Fto; Wo; YY; IT

Outputvector:

Ye u ~

1.5 - uvw. Before the time derivatives of the aircraft's coordinates can be calculated, it is first necessary to compute the body axes velocity components. This is done in the block uuw.Type help uvw a t the MATLABcommand line for on-line help. The structure of uvw is shown in figure F-24.

Inputvector:

[ xT qotTWolTY ~ z lT~ T

Outputvector:

Ybvez

The wind velocity components have to be added to u, v, and w to obtain correct results.

1.6 - x.yHdot.This block is used to calculate the time derivatives of the coordinates of the aircraft x, ,y e , and H (= --z,). Type help xyhdot a t the MATLABcommand line for on-line help. Figure F-25 shows the internal structure of xyHdot.

Inputvector:

[ xT F t 2 Mto? ~ h l ; IT

Outputvector:

y v h

2 - xdotcorr. This block is used to made the implicit P-equation (section 2.2.3) explicit. This correction is valid for the 'Beaver', but if models of other aircraft which also exhibit implicit state equations are implemented in this stucture, it is relatively easy to change xdotcorr. Type help xdotcorr a t the MATLABcommand line for on-line help. The internal structure of this block is shown in figure F-26.

Inputvectors:

x, x (uncorrected), and yat,

Outputvector:

x (corrected to make equation explicit)

The block beta-dot correction contains the following expression:

where the variables b, S, m, and CYbdot have been defined in the Maskdefinitions line:

b = GMl(2); S = GMl(3); m = GMl(10); CYbdot = AM(2,19);

See tables F-1 and F-3 for the definitions of AM and GM1.

h. Other outputs.

Three other outputs blocks have been added to this list, to demonstrate the flexibility of the model. The blocks calculate flightpath variables and accelerations & specific forces, respectively:

fl~ath.This block calculates the flightpath angle y, flightpath acceleration fpa, azimuth angle X , and bank angle 0.Type help flpath a t the MATLAB command line for on-line help. The internal structure of flpath is shown in figure F-27.

Inputvectors:

x a n d x

Outputvector:

Yfi

Accel. This block can be used to calculate accelerations and specific forces (outputs of acceleratometers) in the aircraft's centre of gravity. Type help accel a t the MATLABcommand line for on-line help. The internal structure of accel is shown in figure F-28.

Inputvectors:

Ftot and Fgr

Outputvector:

Yacc

Соседние файлы в предмете [НЕСОРТИРОВАННОЕ]