THOMAS’ ALGORITHM FOR THE SOLUTION OF A TRIDIAGONAL SYSTEM OF EQUATIONS 535

gaussian elimination to the tridiagonal system of equations. Specifically, we

wish to eliminate the lower-diagonal term (the b,’s), as follows. Multiply
Eq. (A.1) by b2.
b2d II -]—bzai>z c b2 (A6)
Multiply Eq. (A.2) by di
d b2 1 +did2 2 *d •2u3 = czdi (A.7)
Subtract Eq. (A.6) from (A.7).
(did2 f›2 ff1) 2 + di ff2 3 ' C2d — lf›2 (A.8)
Divide Eq. (A.8) by di-
f›2a1 c b2
d2 (A.9)

Note that Eq. (A.9) no longer hasa lower-diagonal term—it has been eliminated by
our multiplication and subtraction process above. Let us denote some of the
coefficients in Eq. (A.9) as follows:
f 2ff
l
d 2 = d2 (A.10)
dl

cib2
and C’2— Cz (A.11)
di
Then Eq. (A.9) is written ina simpler form as
d 2u2 a2u c2 (A. 12)
Let us continue with our elimination process by multiplying Eq. (A.12) by b3.
b 3d2u2 +f›3a2 u3 ' b3c2 (A.13)

Multiply Eq. (A.3) by d2.

d 2b u +d 2d3 3+ d 2 a3u4 d 2c3 (A.14)
Subtract Eq. (A.13) from (A.14).
{d 2d3 —b3 a2)u d 2 a3 •d d 2c3 —b3c (A.15)

Divide Eq. (A.15) by d2.

b3C2
3 + ff34 C3 (A.16)
d2 d2
Note that Eq. (A.16) no longer hasa lower-diagonal term; it has been eliminated in
the same fashion as was the case for Eq. (A.9).
536 APPENDIX A

Before we go any further, notice the pattern that is developing here. Equation
(A.9) can be viewed as obtained from Eq. (A.2) by dropping the first term (the term
involving •i). replacing the main-diagonal coefficient with
b2 1
d2 (A.17)
di
instead ofd2. keeping the third term unchanged (a2•3), and replacing the term on
the right-hand side of the equation by
cib2
<2 (A.18)
di
instead ofc2- Comparing Eqs. (A.16) and (A.3), we see exactly the same pattern,
where in Eq. (A.3) the first term is dropped (b3•2). the diagonal coefficient is
replaced by
b3a2
d3 (A,19)
d 2

thethird term remains unchanged (a3•4). and the right-hand side is replaced by
czf›3
C3 — t (A.20)
d2
The pattern is clear. Compare theforms given by (A.17) and (A.19); they are the
same. Compare theforms given by (A.18) and (A.20); they arethe same. Starting at
the top of our system of equations represented by Eqs. (A.1) through (A.5), we leave
Eq. (A.1) alone, but in all the following equations we drop the first term, replace the
coefficient of the main-diagonal term by

d,' —— d, i —— 2, 3, ..., M (A.21)

and replace the term on the right-hand side of the equation by

Ci
ib, i = 2,..., 3, M (A.22)

This will result in an upper bidiagonal form of equations given by

C1
d 2 ’2 “ ‘2 3
d3u3 + a3
u4 C3’

d M _ uM — 1 + M 1 M M 1 (A.23)
d M M — M (A.24)
 THOMAS' ALGORITHM FOR THE SOLUTION OF A TRIDIAGONAL SYSTEM OF EQUATIONS 537

Examining theabove system of equations, we note that the last equation, Eq.
(A.24), contains only one unknown, namely, •M. hence

’M
M (A.25)
dM

The solution of the remaining unknowns is obtained by working upward in the

above system. For example, after ow is obtained from Eq. (A.25), the value ofuM— I
can be found from Eq. (A.23) as
M — 1 ’M
(A.26)
d M —1
Indeed, by inspection we can see that Eq. (A.26) can be replaced by the general
recursion formula

(A.27)

for the calculation of u;, where•i+} has already been calculated from theprevious
application of Eq. (A.27).
In summary, Thomas’ algorithm is as follows. Givena system of linear,
simultaneous, algebraic equations in tridiagonal form represented by Eqs. (A.1) to
(A.5), we first change this system into an upper bidiagonal form by dropping the
first term in each equation (involving the b;’s), replacing the coefficient of the main-
diagonal term by Eq. (A.21), and replacing the right-hand side with Eq. (A.22). This
will result in the last equation in the system in having only one unknown, namely,
•M. Solve foruM *• m Eq. (A.25). Then, all other unknowns arefound in sequence
from Eq. (A.27), starting with u; = up— and ending with u; = u i
For your reference, the computer listing used to solve the Couette flow
problem described in Sec. 9.3 is listed below. This computer program is essentiallya
program forThomas’ algorithm and can be used asa guide to construct your own
computer program for Thomas’ algorithm.

FORTRAN Computer Listing: solution of Couette flow by means of Thomas’

algorithm

REAL U(41),A(41),B(41), D(41), Y(41), C(41)

N=2/
NN=N+1
Y(l)=/./
DEL—1.//FLOAT(N)
RE=5./E+03
EE-1./
TIME=/./
DELTIM=EE*RE*DEL**2
C BOUNDARY CONDITIONS
538 APPENDIX A

U(l)=@.@
U(NN)-1.$
AA=—/.5*EE
BB=1.$+EE
KKEND=2
KKMOD=1
INITIAL CONDITIONS
DOl J=2,N
U(J)=$.$
1 CONTI NUE
A(1)=l.$
B(1)=1./
C(1)=1.$
D(l)=l.$
DO5 KK=1,KKEND
SET ORIGINAL COEFFICIENTS
DO2 J=2,N
Y(J)=Y(J-1)+DEL
A(J)=AA
IF(J.EQ.N) A(J)=$.$
D(J)—BB
B(J)=AA
IF(J.EQ.2) B(J)=/./
C(J)=(1./-EE)*U(J)+/.5*EE*(U(J+1)+U(J’1))
IF(J.EQ.N) C(J)=C(J)—AA*U(NN)
2 CONTINUE
C UPPER BIDIAGONAL FORM
DO3 J=3rN
D(J)=D(J)—B(J)*A(J—1)/D(J—1)
C(J)=C(J)-C(J-l)*B(J)/D(J-l)
CONTINUE
CALCULATION OF U(J)
DO4 K=2,N
M=N-(K-2)
U(M)=(C(M)-A(M)’U(M+1))/D(M)
4 CONTINUE
Y=(l)=$./
Y(NN)=Y(N)+DEL
TIME=TIME+DELTIM
TEST=MOD(KK,KKMOD)
IF(TEST.GT.j.$l) GO TO 5
WRITE(6,1$$) KK, TIME,DELTIM
WRITE(*,1/$) KK,TIME,DELTIM
WRITE(6,1/1)
WRITE(*,1/1)
WRITE(6,1/2) (J,Y(J),U(J),B(J),D(J),A(J),C(J),J=1,NN)
WRITE(’,1/2) (J¿ (J)¿U J)rB J)yD(J)¿A(J),C(J1,J-1,NN)
CONTINUE
FORMAT(5X//5X, ’SOLUTION AT’,5X, ’KK=’,I3,5X, ’TIME=’,E1/.3,5X
’DELTIM=’,El$.3//)
1$1 FORMAT(3X, ’J’,6X, 'Y’,9X, ’U’,9X, ’B’,9X, ’D’,9X, ’A’,9X, ’C’)
1$2 FORMAT(2X,I3,6E1j.3)
ENDӄ
 REFERENCES

1. Anderson, John D., Jr.: Introduction to Flight, 3d ed., McGraw-Hill, New York, 1989.
2. Anderson, John D., Jr.: H ypersonic and High Temperature Gas Dynamic's McGraw-Hill, New
York, 1989.
3. Rouse, Hunter, and Simon Ince: History of Hydraulics, Iowa Institute of Hydraulic Research,
Annes, Iowa 1957.
4. Tokaty, G. A.:A History and Philosophy of Fluid Mechanics, G. T. Foulis, Henly-on-Thames,
England, 1971.
5. Anderson, John D., Jr.: The History of Aerodynamics, and Its Impact on Flying Machines,
Cambridge University Press, New York (in preparation).
6. Kothari, A. P., and J. D. Anderson, Jr.: “Flows Over Low Reynolds Number Airfoils—Com-
pressible Navier—Stokes Numerical Solutions,” AIAA paper 85-0107, presented at AIAA 23rd
Aerospace Sciences Meeting, Reno, Nev., Jan. 1W17, 1985.
7. Pohlen, L. J., and T. J. Mueller: “Boundary Layer Characteristics of the Miley Airfoil at Low
Reynolds Numbers,” J. Aircr., vol. 21, no. 9, pp. 658—664, September 1984.
8. Anderson, John D.,Jr.: Fundamentals of Aerodynamics, 2d ed., McGraw-Hill, New York, 1991.
9. Bush, Richard J., Jr., Merle Jager, and Brad Bergman: “The Application of Computational Fluid
Dynamics to Aircraft Design,” AIAA paper 86-2651, 1986.
10. Jameson, A., W Schmidt, and E. Turkel: “Numerical Solutions of the Euler Equations by Finite
Volume Methods Using Runge-Kutta Time Stepping Schemes,” AIAA paper 81-1259, 198l.
11. Chapman, Dean R.:“Computational Aerodynamics Development and Outlook,” AIAA J., vol. 17,
no. 12, pp. 1293-1313, December 1979.
12. Moretti, G., and M. Abbett: “A Time-Dependent Computational Method forBlunt Body Flows,”
AIAA J., vol. 4, no. 12, pp. 2136-2141, December 1966.
13. Anderson, Dale A., John C. Tannehill, and Richard H. Pletcher: Computational Fluid Mechanics
and Heat Transfer, McGraw-Hill, New York, 1984.
14. Fletcher, C. A.: Computational Techniques for Fluid Dynamics, vol. I: Fundamental and General
Techniques, Springer-Verlag, Berlin, 1988.
15. Fletcher, C. A.: Computational Techniques for Fluid Dynamics, vol. II: Specific Techniques %r
Different Flow Categories, Springer-Verlag, Berlin, 1988.
16. Hirsch, Charles: Numerical Computation ofInternal and External Flows, vol. 1: Fundamentals of
Numerical Discretization, Wiley, New York, 1988.
17. Hirsch, Charles: Numerical Computation of Internal and External Flow's, vol. II: Computational
Methods for Inviscid and Viscous Flows, Wiley, New York, 1990.
18. Hoffmann, K. A.: Computational Fluid DynamfCS for Engineers, Engineering Education System,
Austin, Tex., 1989.
19. Hildebrand, Francis B.: Advanced Calculus %r Applicafions, 2d ed., Prentice-Hall, Englewood
Cliffs, N.J., 1976.
20. Schlichting, H.: Boundars Lai'er Theotn, 7th ed., McGraw-Hill, New York, 1979.
21. Anderson, John D.,Jr.: Modein Compressible Flow.’ With Historical Perspective, 2d ed., McGraw-
Hill, New York, 1990.
22. Kreyszig, E.: Advanced Engineering Mathematics, Wiley, New York, 1962.
23. Whitham, G. B.: Linear and Nonlinear Waves, Wiley, New York, 1974.
24. Annes, W. F.: Nonlinear Partial Diff'erential Equations inEngineering, Academic, New York, 1965.
25. Courant, R., K. O. Friedrichs, and H. Lewy: “Uber die Differenzengleichungen der Mathema-
tischen Physik,” Math. Ann, vol. 100, p. 32, 1928.
26. Thompson, Joe F. (ed.): Numerical Grid Generation, North-Holland, New York, 1982.

539
540 REFERENCES

27. Thompson, JoeF., Z. V A. Warsi, and C. Wayne Mastin: Numerical Grid Generation. Foundations
and Applications, North-Holland, New York, 1985.
28. Viviand, H.: “Conservative Forms of Gas Dynamic Equations,” Rech. Aero.sp., no. 1971-1, pp.
65-68, 1974.
29. Vinokur, M.: “Conservation Equations of Gas Dynamics in Curvilinear Coordinate Systems,” J.
Comput. Phys., vol. 14, pp. 105-125, 1974.
30. Mullins, G. A., J. D. Anderson, Jr., and J. P. Drummond: “Numerical Investigation of Supersonic
Base Flow with Parallel Injection,” AIAA paper 82-1002, 1982.
31. Mullins, G. A.: “Numerical Investigation of supersonic Base Flow with Tangential Injection,” M.S.
thesis, Department of Aerospace Engineering, University of Maryland, College Park, 1981.
32. Holst, T. L.: “Numerical Solution of Axisymmetric Boattail Fields with Plume Simulators,” AIAA
paper 77-224, 1977.
33. Roberts, B. O.: “Computational Meshes forBoundary Layer Problems,” Lecture Notes inPhysics,
Springer-Verlag, New York, pp. 171-177, 1971.
34. Thompson, J. F., F. C. Thames, and C. W. Mastin: “Automatic Numerical Generation of Body-
Fitted Curvilinear Coordinate Systems for Fields Containing Any Number of Arbitrary Two-
Dimensional Bodies,” J. Comput. Phys., vol. 15, pp. 299-319, 1974.
35. Corda, Stephen: “Numerical Investigation of the Laminar, Supersonic Flow over a Rearward-
Facing Step Using an Adaptive Grid Scheme,” M.S. thesis, Department of Aerospace Engineering,
University of Maryland, College Park, 1982.
36. Dwyer, H. A., R. J. Kee, and B. R. Sanders: “An Adaptive Grid Method forProblems in Fluid
Mechanics and Heat Transfer,” AIAA paper 79-1464, 1979.
37. Steinbrenner, John P., and Dale A. Anderson: “Grid-Generation Methodology in Applied
Aerodynamics,” in P. A. Henne (ed.), Applied Computational Aerodynamics Progress in Astro-
nautics and Aeronautics Series, vol. 125, AIAA, Washington, D.C., chap. 4, pp. 91-130, 1990.
38. Karman, S. L., Jr., J. P. Steinbrenner, and K. M. Kisielewski: “Analysis of the F-16 Flow Field bya
Block Grid Euler Approach,” AGARD Conf. Proc. 412, 1986.
39. Venkatakrishnan, V, and D. J. Mavriplis: “Implicit Solvers for Unstructured Meshes,” AIAA
paper 91-1537-CP, Proc. AIAA 10th Comput. Fluid Dyn. Conf., pp. 115-124, June 24—27,
1991.
40. Hassan, O., K. Morgan, J. Peraire, E. J. Probert, and R. R. Thareja: “Adaptive Unstructured Mesh
Methods forSteady Viscous Flow,” AIAA paper 91-1538-CP, Proc. AIAA 10th Comput. FluidDyn.
Conf., pp. 125-133, June 24-27, 1991.
41. DeZeeuw, Darren, and Kenneth G. Powell: “An Adaptively-Refined Cartesian Mesh Solver for
the Euler Equations,” AIAA paper 91-1542-CP, Proc. AIAA 10th Comput. Fluid Dyn. Conf., pp.
166-180, June 24-27, 1991.
42. Rubbert, Paul, and Dockan Kwak (eds): AIAA 10th Computational Fluid Dynamics Conference,
June 24—27, 1991.
43. MacCormack, R, W.: “The Effect of Viscosity in Hypervelocity Impact Cratering,” AIAA paper
69-354, 1969.
44. Kuruvila, G., and J. D. Anderson, Jr.: “A Study of the Effects of Numerical Dissipation on the
Calculation of Supersonic Separated Flows,” AIAA paper 85-0301, 1985.
45. Annes Research Staff: “Equations, Tablgs, and Charts for Compressible Flow,” NACA Rep. 1135,
1953.
46. Abbett, M. J.: “Boundary Condition Calculation Procedures forlnviscid Supersonic Flow Fields,”
Proc. 1st AIAA Comput. Fluid Dyn. Conf., pp. 153 172, 1973.
47. Shang, J. S., and S. J. Scherr: “Navier—Stokes Solutions fora Complete Re-Entry Configuration,”
J. Aircr., vol. 23, no. 12, pp. 881-888, December 1986.
48. Schroder, W., and F. Mergler: “Comparative Study of Inviscid and Viscous Flows Over an STS,” in
C. Hirsch, J. Periaux, and W. Kordulla (eds.), Computational Fluid Dynamics ’92, vo1. 1, Elsevier,
Amsterdam, 1992, pp. 323-330.
49. Turkel, E., R. C. Swanson,V N. Vatsa, and J. A. White: “Multigrid for Hypersonic Viscous Two-
and Three-Dimensional Flows,” AIAA paper 91-1572-CP, Proc. AIAA 10th Comput. Fluid Dyn.
Conf., 1991.
50. Vilsmeier, R., and D. Hanel: “Adaptive Solutions for Compressible Flows on Unstructured,
 REFERENC ES 541

Strongly Anisotropic Grids,” in C. Hirsch, J. Periaux, and W Kordulla (eds.), Computational Fluid
Dynamics ’92, vol. 2, Elsevier, Amsterdam, 1992, pp. 945-951.
51. Vassberg, J. C., and K. R. Dailey: “AIRPLANE: Experiences, Benchmarks and Improvements,”
AIAA paper 90-2998, 1990.
52. Petot, B., and A. Fourmaux: “Validation of Viscous and lnviscid Computational Methods Around
Axial Flow Turbine Blades,” in C. Hirsch, J. Periaux, and W. Kordulla (eds.), Computational Fluid
Dynamics ’92, vol. 2, Elsevier, Amsterdam, 1992, pp. 611-618.
53. Enomoto, S., and C. Arakawa: “2-D and3 -D Numerical Simulation of a Supersonic Inlet
Flowfield,” in C. Hirsch, J. Periaux, and W. Kordulla (eds.), Computational Fluid Dynamics ’92,
vol. 2, Elsevier, Amsterdam, 1992, pp. 781-788.
54. Borland, C. J.: “A Multidisciplinary Approach to Aeroelastic Analysis,” in A. K. Noor and
S.L. Venneri (eds.), Computing Systems in Engineering, vol. 1, Pergamon, New York, 1990, pp.
197-209.
55. Vandromme, D., and A. Saouab: “Implicit Solution of Reynolds-Averaged Navier—Stokes
Equations for Supersonic Jets on Adaptive Mesh,” in C. Hirsch, J. Periaux, and W. Kordulla
(eds.), Computational Fluid Dynamics ’92, vol. 2, Elsevier, New York, 1992. pp. 727—731.
56. Kandil, O. A., H. A. Kandil, and C. H. Liu: “Supersonic Quasi-Axisymmetric Vortex Breakdown,”
AIAA paper 91-3311-CP, Proc. AIAA 9thAppl. Aerodyn. Conf., pp. 851 —863, 1991.
57. Rai, M. M., and P. Moin: “Direct Numerical Simulation of Transition and Turbulence ina Spatially
Evolving Boundary Layer,” AIAA paper 91-1607-CP, Proc. AIAA 10th Comput. Fluid H -n. Coiif.,
pp. 890-914, 1991.
58. Shaw, C. T.: “Predicting Vehicle Aerodynamics Using Computational Fluid Dynamics—A User's
Perspective,” Research in Automotive Aerodynamics, SAE Special Publication 747. pp. 119-132,
February 1988.
59. Matsunaga, K., H. Mijata, K. Aoki, and M. Zhu: “Finite-Difference Simulation of 3D Vortical
Flows Past Road Vehicles,” Vehicle Aerodynamics, SAE Special Publication 908, pp. 65-84,
February 1992.
60. Griffin, M. E., R. Diwaker, J. D. Anderson, and E. Jones: “Computational Fluid Dynamics Applied
to Flows in an Internal Combustion Engine,” AIAA paper 78-57, presented at AIAA 1 6th
Aerospace Sciences Meeting, January 1978.
61. Mampaey, F., and Z. A. Xu: “An Experimental and Simulation Study ofa Mould Filling Combined
with Heat Transfer,” in C. Hirsch, J. Periaux, and W. Kordulla (eds. ). Computational Fluid
Dynamics ’92, vol. 1, Elsevier, Amsterdam, 1992, pp. 421—428.
62. Steijsiger, C., A. M. Lankhorst, and Y. R. Roman: “Influence of Gas Phase Reactions on the
Deposition Rate of Silicon Carbide from the Precursors Methyltrichlorosilane and Hydrogen,” in
C. Hirsch, O. C. Zienkiewicz, and E. Onate (eds.), Numerical Methods in Engineering ’92,
Elsevier, Amsterdam, 1992, pp. 857-864.
63. Toorman, E. A.,and J. E. Berlamont: “Free Surface Flow ofa Dense, Natural Cohesive Sediment
Suspension,” in C. Hirsch, I. Periaux, and W. Kordulla (eds.), Computational Fluid Dv'namics ’92,
vol. 2, Elsevier, Amsterdam, pp, 1005-1011, 1992.
64. Bai, X. S., and L. Fuchs: “Numerical Model forTurbulent Diffusion Flames with Applications,” in
C. Hirsch, J. Periaux and W. Kordulla (eds.), Computational Fluid fit’narii ics ’92, vol. 1, Elsevier,
Amsterdam, 1992, pp. 1 69—176.
65. McGuirk, J. J., and G. E. Whittle: “Calculation of Buoyant Air movement inBuildings—Proposals
fora Numerical Benchmark Test Case, Computational Fluid H'narnicefor theEnvironmental and
Building Services Engineer—Tool or To›? The Institution of Mechanical Engineers, London, pp.
13-32, November 1991.
66. Alamdari, F., S. C. Edwards, and S. P. Hammond: Microclimate Performance of an Open Atrium
Office Building: A Case Study in Thermo-Fluid Modeling.” Computational Fluid Dynamics for
theEnvironmental and Building Screices Engineet —Tool or Top'? The Institution of Mechanical
Engineers, London, pp. 81—92, November 1991.
67. Patankar, S. V, and D. B. Spalding: “A Calculation Procedure for Heat, Mass and Momentum
Transfer in Three-Dimensional Parabolic Flows,” Int. J. Heat Mass Transfer, vol. 15, pp. 1787—
1806, 1972.
68. Patankar, S. V: Numerical Heat Transfer and Fluid Flow, Hemisphere, New York, 1980.
542 REFERENCES

69.Oran, Elaine S., and Jay P. Boris: Numerical Simulation o[Reactive Flow, Elsevier, New York,
1987.
70. Jacquotte, 0. P., and G. Coussement: “Structural Grid Variation Adaption: Reaching theLimit?” in
C. Hirsch, J. Periaux, and W. Kordulla (eds.), Computational Fluid Dynamics ’92, vol. 2, Elsevier,
Amsterdam, 1992, pp. 1077-1087.
71. Degani, D, and Y. Levy: “Asymmetric Turbulent Vortical Flows over Slender Bodies,” Proc. AIAA
9thAppl. Aerodyn. Conf., pp. 756 765, September 1991.
72. TECPLOT Users Manual, version 5, Amtec Engineering, Inc., Bellevue, Wash., 1992.
73. Selmin, V, E. Hettena, and L. Formaggia: “An Unstructured Node Centered Scheme forthe
Simulation of 3-D Inviscid Flows,” in C. Hirsch, J. Periaux, and W. Kordulla (eds.), Computational
Fluid Dynamics ’92, vol. 2, Elsevier, Amsterdam, 1992, pp. 823-828.
74. MacCormack, R. W : “Current Status of Numerical Solutions of the Navier—Stokes Equations,”
AIAA paper 88-0513, 1988.
75. Van Driest, E. R.: “Investigation of Laminar Boundary Layer in Compressible Fluids Using the
Crocco Method,” NACA Tech. Note 25 79, January 1952.
76. Stollery, J. L.: “Viscous Interaction Effects and Re-entry Aerothermodynamics: Theory and
Experimental Results,” Aerodynamic Problems of Hypersonic Vehicles, vol. 1, AGARD Lecture
Series 42, pp. 10-1-10-28, July 1972.
77. Beam, R. M., and R. F. Warming: “An Implicit Finite-Difference Algorithm for Hyperbolic
Systems in Conservation Law Form,” J. Comput. Phys., vol. 22, pp. 87-110, 1976.
78. Briley, W. R., and H. McDonald: “Solution of the Three-Dimensional Navier—Stokes Equations by
an Implicit Technique,” Proceedings o the Fourth International Conference on Numerical
Methods in Fluid Dynamics, Lecture Notes in Physics, vol. 35, Springer-Verlag, Berlin, 1975.
79. Peaceman, D. W., and H. H. Rackford: “The Numerical Solution of Parabolic and Elliptic
Differential équations,” J. Soc. Ind. Appl. Math., vol. 3, pp. 28W1, 1955.
80. Douglas, J., and H. H. Rackford: “On the Numerical Solution of Heat Conduction Problems inTwo
and Three Space Variables,” Trans. Am. Math. Soc., vol. 82, pp. 4231W239, 1956.
81. Godunov, S. K.: “A Difference Scheme forNumerical Computation of Discontinuous Solution of
Hydrodynamic Equations,” Math. Sb., vol. 47, pp. 271-306, 1959, in Russian; translated U.S. Joint
Publications Research Service, JPRS 7226, 1969.
 INDEX

Adaptive grid (see Grids) Caloric equation of state, 79

ADI {see Alternating-direction-implicit Cartesian grid (see Grids)
technique) Cell Reynolds number, 456—457
Adiabatie wall condition, 81, 468 Central difference (see Finite differences)
Adiabatic wall temperature, 81 CFD [see Computational fluid dynamics)
Aeroelastieity, 526 CFD-generated schlieren, 270, 271
Aircraft flowfields: CFL condition (see Courant-Friedrichs-Levy
generic fighter, 279 condition)
hypersonie body, 274, 278 Characteristic lines, 97, 99, 162, 488
Northrop F-20, 10-13, 209 Civil engineering applications, 19, 20, 22
space transportation system, 519, 520 Compatibility equation, 101
X-24C, 517, 518 Composite plots, 274, 278
Alternating-direction-implicit (ADI) technique, Computational costs, 27
243-247, 494—495 Computational 8uid dynamics:
Amplification factor, 1 60, 165 definition of, 23, 25, 26
Analytical domain, 163, 164 as design tool, 9—13
Approximate factorization, 247, 492W96 new approach in, 2—3
Approximate Riemann solver, 499, 507, 528 as research tool, 6-9
Artificial viscosity, 236, 238-243 Computational plane, definition of, 170, 171
forMacCormack's technique, 238, 363-364, Computer graphic techniques, 264-279, 516-517
366—370 Computer programming (see Programming
Automobile flowfields, 14-17 procedures)
Conservation form:
Backward difference [see Finite differences) of continuity equations, 51, 55
Base flow, 191 of energy equation, 74
(See also rearward-facing step) general discussion of, 42, 90, 225, 480W82
Beam-Warming method {see Implicit methods) generic form of, 83, 481
Block tridiagonal matrices, 49W97 of momentum equation, 65, 66
Blunt body, supersonic, 29, 30, 119, 120 for quasi-one-dimensional nozzle 8ow, 336—
Body forces, 61 356
Boundary conditions: strong form of, 88, 183, 185, 377, 452
Abbett’s condition (inviscid flow over walls), transformed form of, 185
392—395 weak form of, 88
forconservation form, 343 Consistent equation (see Finite differences)
no-slip, 80, 457 Continuity equation, 49—60
physical, 90-92, 392 differential form, 55, 56
for pressure correction method 262, 263, 437 integral form, 51, 53
rejection, 138 one-dimensional 8ow, 482
for subsonic inflow, 303-306 quasi-one-dimensional flow, 286, 291
forsubsonic outflow, 327-328 Contour plots:
forsupersonic out8ow, 305—307 density, 272, 520, 525
Boundary-fitted grid [see Grids) flooded, 266, 268
Boundary layer 8ows, 113, 114, 450, 472 general discussion of, 265—270, 516

543
544 INDEX

Contour plots (Cont.): Errors {Cont.):

gray-scale color, 266 high-frequency, 513
Mach number, 523, 524, 527 low-frequency, 513
pressure, 11, 241, 272, 279, 521, 524 round-off, 155
velocity, 267, 268 Euler equations, 77-79, 154
vorticity, 15, 530-533 Euler explicit form, 162
Control surface, 41 Explicit methods, general discussion of, 145-153
Control volume, 41 (See also Lax-Wendroff technique;
Convective derivative, 45 MacCormack's technique)
Couette flow, 41W45
Courant-Friedrichs-Levy (CFL) condition, 162,
302, 395, 457 Finite differences:
Courant number, 162, 324-325 based on Taylor's series, 128
Cramer's rule, 98, 178, 230 consistent difference equation, 144
Crank-Nicolson method (see Implicit methods) difference equations, 142—145
CRAY computers, 28, 533 first-order forward, 130
first-order rearward, 131
fourth-order central, 135
Delta form, 496 general concept, 123, 127
Detonation wave, 266-268 modules, 134-136, 147, 149, 494
Diagonalization, 500 one-sided, 139
Difference equation (see Finite differences) second-order central, 132
Direct metrics [see Metrics) second-order second central, 132-134
Direct numerical simulation, 530-533 upwind, 499
Dirichlet condition, 118 Finite volumes:
Discretization, 125, 165 discretized equations, 167
Dispersion, 237 general concept, 123
(See also Numerical dissipation) Flat plate flow, 447—476
Dissipation (see Numerical dissipation) Flowcharts, 459—466
Divergence form, 79 (See also Programming procedures)
Divergence of velocity, 47, 48 Fluid element model, 41, 42
Domain ofdependence, 106, 107 Flux-corrected transport (FCT) method, 266
Flux limiters, 509, 510, 512
Flux terms, 84, 185, 339, 341, 380
Eigenvalues, 102, 482, 487—489 Flux variables, 85
Eigenvector, 500 Flux-vector splitting, 500-502, 510, 511
Elliptic nature: Forward difference (see Finite differences)
definition of, 100, 103 Furnace applications, 21
equations, 30, 104, 105
general discussion of, 117—119
of pressure correction formula, 160 Gauss-Seidel method, 231
regions in flow, 29, 277 Godunov schemes, 499, 502—508
Energy equation, 66—74 Governing 8ow equations:
differential form, 70, 71 generic form, 83
one-dimensional flow, 482 introduction to, 38W0
quasi-one-dimensional flow, 286, 295, 296 forquasi-one-dimensional flow, 296
Engine calculations, 14, 16, 17 summary of, 75-80
Engineering Research Center for Computational transformed generic flow, 185
Field Simulation, 276 Grid independence, 322—324, 355
Environmental engineering applications, 20—23, Grid points, 126, 137, 299, 423
25, 26 Grids:
Errors: adaptive, 200—208
boundary condition, 321-322 boundary-fitted, 15, 18, 170, 192-200, 269
discretization, 154, 155 C-type, 194
general discussion of, 153-165 compressed, 15, 186-192
 INDEX 545

Grids {Cont.): Manufacturing applications, 17-19

elliptically generated, 194-200 Marching solutions:
finite volume, 20 general, 14d, 153
generation of, 124, 168, 171 space marching, 225—232, 375
O-type, 194 time marching, 30, 85, 119, 146, 221
rectangular (cartesian), 16, 169, 212—214, 240 Marching variables, 143
staggered, 250—253, 436 Mass source term in pressure correction method,
structured, 126, 210 260, 443, 444
unstructured, 126, 210-212, 523 Mathematical behavior of equations, 95-121, 277
Mesh {see Grids)
Mesh plots, 273, 275—277
Heat conduction equation, 116, 121, 142, 145 {See also Computer graphic techniques)
High-resolution schemes, 508-510 Method of characteristics, 102
Hyperbolic nature: Metrics, 173, 178-183, 206, 207
definitiori of, 100, 103 Miley airfoil, 198, 199
from eigenvalues, 488 Models of theflow, 40—42
equations, 30, 104, 105 Modified equation, 235
general discussion of, 106—111, 416 Momentum equation:
regions in flow, 29, 277 differential form, 64
general discussion of, 60-66
one-dimensiona1 flow, 482
Implicit methods: quasi-one-dimensional flow, 286, 294
Beam-Warming method, 49H492, 497 Monotone variation, 499
Crank-Nicolson method, 148-151, 244, 42a Multigrid method, 513—514, 521
425, 489, 491
general discussion of, 145-153, 489
lagging coeffieients method, 490 Naval architecture applications, 22, 23, 26
linearization, 49 92 Navier-Stokes equations, 64, 66, 75-77, 79, 154,
Initial conditions, 307—308, 344, 362, 420 225, 236, 239, 249, 250, 266, 417, 450, 451,
Initial data lines, 108, 226, 227, 379, 386, 387 490
Inverse metrics, 183 Neumann condition, 118
Inverse transformation {see Transformations) Nonconservation form:
of continuity equation, 53, 56
of energy equation, 70, 72
Jacobi method, 231 general discussion of, 42
Jacobian: of momentum equation, 64
determinant of the transformation, 179, 180, Normal shock wave, 91, 357, 359
206 Nozzle flow {see Quasi-one-dimensional nozzle
of the flux vector, 481, 493 flows)
for one-dimensional flow, 486 Numerical dispersion, 237
Numerical dissipation, 232-243
Numerical domain, 163, 164
Laminar flow, 7-9
Laplace's equation, 121, 176
Lax-Wendroff technique, 217-221 One-dimensional flow, 482W89
Local time stepping, 302-303

Parabolic equations:
MacCormack's technique, 222—229, 238, 288, boundary-layer equations, 113
330, 336, 375, 387, 448, 449, 453W55, 460, for Couette flow, 417, 421
461, 463—465, 474, 497, 517 definition of, 100
Mach angle, 376 general discussion of, 111-117
Mach disk, 528 heat conduction equation, 116
Mach number profiles, flat plate flow, 474, 476 parabolized Navier-Stokes equations, 115
Mach wave, 376 regions in flow, 277
546 INDEX

Parabolized Navier-Stokes equations (see Step size:

Parabolie equations) spatial, 395—397
Parallel processors (see Processors) time, 301-303, 455W57
Particle paths, 15 Streamlines, 21, 271-274
Physical plane, definition of, 170, 171 Structured grid {see Grids)
Point-iterative method, 229 Submarine flow field, 27
Poisson equation, 260 Substantial derivative, 43W6
Prandtl-Meyer expansion wave, 374W15 Successive overrelaxation, 231
Prandtl-Meyer function, 377 Supersonic nozzle flow, two-dimensional, 527
Pressure contours {see Contour plots)
Pressure correction formula, 260, 441
Pressure correction technique, 247—264, 435W45 Taylor's series, 128
Pressure profiles, 469, 475 TECPLOT, 264
Primitive variables, 85, 340, 380, 490 Temperature profiles, 471, 472
Processors: Thermal difhisivity, 116
parallel, 28, 153 Thermal equation of state, 79
vector, 28 Thomas’ algorithm, 150, 243, 245, 246, 424,
Programming procedures, 459W67 426—429, 534-538
Time marching see Marching solutions)
Time step calculation, 301-303, 455W57
Quasi-one-dimensional nozzle flows: (See also Courant-Friedrichs-Levy condition)
general discussion of, 283—372 Total-variation-diminishing schemes, 499, 509—
with shock wave, 356-372 510, 512
subsonic isentropie fiow, 325—336 Transformations:
subsonic-supersonic isentropie flow, 285-325, of first derivatives, 173
336—356 general discussion of, 124, 171-178
inverse, 178
of second derivative, 175, 176
Trapezoidal rule, 491, 493
Rearward-facing step, 240-243 Tridiagonal matrix, 150, 424, 496
{See also Base flow) Truncation error:
Rectangular grid (see Grids) of difference equation, 144
Region of influence, 106, 107, 110 of finite difference, 130
Relaxation technique, 229-232 for grid independence, 322
Residual, 316, 317 Turbulent flow, 7-9
Riemann problem, 504 TVD schemes (see Total-variation-diminishing
Runge-Kutta scheme, 521 schemes)

Scatter plots, 273, 275 Unconditional instability, 162

(See ofso Computer graphic techniques) Unconditional stability, 151, 425
Schlieren (see CFD-generated schlieren) Underrelaxation, 233
Shock-capturing method, 89 92, 356 372 Unstructured grid (see Grids)
Shock-fitting method, 89 Upwind schemes, 497—508
Shock interaction, 268-270 Upwinding (see Upwind schemes)
Shock layer, 449
Shock tube problem, 502-504,5 11, 512
SIMPLE algorithms, 248, 261-262 Van Leers flux splitting, 502
(See also Pressure correction technique) Vector plots, 270—274
Solutions vector, 84, 87, 340 (See also Computer graphic techniques)
Source term, 84 Vector processors (see Processors)
Space marching {see Marching solutions) Velocity profiles:
Stability criterion, 151 for flat plate flow, 473, 475
Stability of solutions, 153—165 for rearward-facing step, 242
Staggered grid (.see Grids) for unsteady Couette flow, 431
 INDEX 547

Viscous interaction, 470 Wave number, 156-158

von Neumann stability method, 161 Well-posed problems, 120
Vortex-shock interaction, 528, 529

Wave diagram, 504, 505 xy plots, 264, 265

Wave equation, 121, 161, 498 (See also Computer graphic techniques)