essential features of moving particle with pressure mesh€¦ · mppm moving particle method with...
TRANSCRIPT
1
Essential Features of Moving Particle with Pressure Mesh
黃耀新 (Yao-Hsin Hwang) 國立高雄海洋科技大學 輪機工程系 National Kaohsiung Marine University Department of Marine Engineering
2
Merits of particle method
Proceeds without topological connection among computational nodes (meshless)
Lagrangean treatment on convection term Easy particle manipulations (addition and/or
deletion) Versatility in engineering problems
3
MPPM
Moving particle method with embedded pressure mesh (MPPM) –
Incompressible Flow Computations
4
Features of existing schemes
SPH, MPS, FVPM…. Material particles (mass point, global mass
conservation) Operators realized with randomly particle
cloud Artificial compressibility or projection
method for continuity constraint
5
Disadvantages of existing schemes
Inaccurate (inconsistent) operator realization (particle smoothing)
Incommensurate particle distribution (invariable length scale)
Incapable particle management Assignment of boundary condition Computationally inefficient (Pressure
equation)
6
Motivation: role of pressure
Continuity constraints – Langragian or Eulerian?
Governing equation without convection term Two types of computational particles:
Langragian velocity/Eulerian pressure Velocity particle liberated from mass
constraint
7
Essentials of present proposition (1)
Inserted pressure mesh to realize pressure-related operators
Background mesh and length scale Projection method for continuity constraint Local mass conservation Particle as observation point
8
Essentials of present proposition (2)
No particle management constraint Feasible non-uniform distribution No special boundary treatment Constant and diagonally dominant
coefficient matrix of pressure equation
9
Governing equations
Continuity
Momentum
0∇ ⋅ =u
2D pDt
∇= − + ν∇
ρu u
10
Projection method
1. Intermediate step
2. Pressure step
3. Final step
* n 2 np p pt ,= + ∆ ν∇u u u * n *
p p pt= + ∆r r u
2 n 1 *pt
+ ρ∇ = ∇ ⋅
∆u
n 1pn 1 *
p p
pt ,
++ ∇
= − ∆ρ
u u 2
n 1 * n 1p p p
t p+ +∆= − ∇
ρr r
11
Pressure mesh Continuity
Facial velocity splitting
n 1 n 1 n 1 n 1e w n s(u u ) y (v v ) x 0+ + + +− ∆ + − ∆ =
n 1 n 1n 1 * E Pe e
t p pu ux
+ ++ ∆ −
= −ρ ∆
12
Pressure equation
Facial velocity interpolation
n 1 n 1 n 1 n 1P E N W
n 1 * * * *S e w n s
x y y y x2( )p p p py x x x yx p [(u u ) y (v v ) x]y t
+ + + +
+
∆ ∆ ∆ ∆ ∆+ = + +
∆ ∆ ∆ ∆ ∆∆ ρ
+ − − ∆ + − ∆∆ ∆
* *e p pe e pe e
p pu u (r , r ) / (r , r )= ω ω∑ ∑
13
Pressure gradient
Shape function
P E
N NE
p( , ) p (1 )(1 ) p (1 )p (1 ) p
ξ η = − ξ − η + ξ − η+ − ξ η + ξη
P E P(x x ) /(x x )ξ = − −
P N P(y y ) /(y y )η = − −
14
Velocity Diffusion
Particle smoothing
j i ji22i
j iji jij i
( )4r≠
≠
φ − φ ω∇ φ =
ω ∑∑
ee
e
r r r1(r, r ) r
otherwise0
≤−ω =
15
Pure convection
x
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
φ
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1UDSKUD2UDQUICKparticle method
θ=15o
x
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
φ
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1UDSKUD2UDQUICKparticle method
θ=30o
x
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
φ
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1UDSKUD2UDQUICKparticle method
θ=45o
16
Pure convection
x
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
φ
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
1UDSKUD2UDQUICKparticle method
17
Lid-driven cavity
Particle velocity Particle distribution
Streamlines at various computation times Reference streamlines
18
Lid-driven cavity
Re=1000
19
Lid-driven cavity
Recirculation flow rate
20 Vertical velocity
Horizontal velocity
u
y
0.0
0.2
0.4
0.6
0.8
1.0
nC=40nC=80Ghia et al.Erturk et al.
0.0 0.0 0.0 0.0 0.0 0.0-0.5 0.5 1.0
Re=10
0
Re=4
00
Re=1
000
Re=2
500
Re=5
000
Re=1
0000
x0.0 0.2 0.4 0.6 0.8 1.0
v
nC=40nC=80Ghia et al.Erturk et al.
Re=100
Re=5000
Re=2500
Re=1000
Re=400
Re=10000
-1.0
0.0
0.0
0.0
0.0
0.0
0.5
-0.5
0.0
Lid-driven cavity
21
Weird consequences
Pronounced solution variations in low-Reynolds number flows
Inaccurate results with denser particle distribution
Inaccurate diffusion operators?
22
Analysis of Laplacian operator
j i ji j i jix2 2
j i j iji ji ji jij i j i
2j i ji j i ji
y xx2 2j i j iji ji
2j i j i ji j i ji
xy yy2 2j i j iji ji
( ) (x x )4 4 [r r
(y y ) (x x )1r 2 r
(x x )(y y ) (y y )1 ]r 2 r
≠ ≠≠ ≠
≠ ≠
≠ ≠
φ − φ ω − ω≈ φ
ω ω
− ω − ω+φ + φ
− − ω − ω+φ + φ
∑ ∑∑ ∑
∑ ∑
∑ ∑
Inconsistency: O(1/ )δ
23
Smoothing difference
gradient model
2 3 2q p q p q p qp q p q p qp q p qp q p q p qp
qp x y xx xy2 2 2 2q p q p q p q p q pqp qp qp qp qp qp
2q p q p qp
yy 2q p qp
x x (x x ) (x x )(y y ) (x x ) (x x ) (y y )1( )r r r r 2 r r
(x x )(y y )1 HOT2 r
≠ ≠ ≠ ≠ ≠
≠
φ − φ − − ω − − ω − ω − − ωω = φ + φ + φ + φ
− − ω+ φ +
∑ ∑ ∑ ∑ ∑
∑
2 2q p q p q p q p qp q p qp q p q p qp
qp x y xx2 2 2q p q p q p q pqp qp qp qp qp
2 3q p q p qp q p qp
xy yy2 2q p q pqp qp
y y (x x )(y y ) (y y ) (x x ) (y y )1( )r r r r 2 r
(x x )(y y ) (y y )1 HOTr 2 r
≠ ≠ ≠ ≠
≠ ≠
φ − φ − − − ω − ω − − ωω = φ + φ + φ
− − ω − ω+φ + φ +
∑ ∑ ∑ ∑
∑ ∑
24
Smoothing difference
Laplacian model 3 2 4 3
q p q p q p qp q p q p qp q p qp q p q p qp2qp x y xx xy2 4 4 4 4
q p q p q p q p q pqp qp qp qp qp qp
2 2q p q p qp
yy 4q p qp
x x (x x ) (x x ) (y y ) (x x ) (x x ) (y y )1( )r r r r 2 r r
(x x ) (y y )1 HOT2 r
≠ ≠ ≠ ≠ ≠
≠
φ − φ − − ω − − ω − ω − − ωω = φ + φ + φ + φ
− − ω+ φ +
∑ ∑ ∑ ∑ ∑
∑
2 2q p q p q p q p q p qp q p q p qp
qp x y2 4 4q p q p q pqp qp qp qp qp
3 2 2q p q p qp q p q p qp
xx xy4 4q p q pqp qp
3q p q p qp
yy 4q p qp
x x y y (x x ) (y y ) (x x )(y y )( )( )
r r r r r
(x x ) (y y ) (x x ) (y y )12 r r
(x x )(y y )1 HOT2 r
≠ ≠ ≠
≠ ≠
≠
φ − φ − − − − ω − − ωω = φ + φ
− − ω − − ω+ φ + φ
− − ω+ φ +
∑ ∑ ∑
∑ ∑
∑
2 3 2 2q p q p q p q p qp q p qp q p q p qp2
qp x y xx2 4 4 4q p q p q p q pqp qp qp qp qp
3 4q p q p qp q p qp
xy yy4 4q p q pqp qp
y y (x x )(y y ) (y y ) (x x ) (y y )1( )r r r r 2 r
(x x )(y y ) (y y )1 HOTr 2 r
≠ ≠ ≠ ≠
≠ ≠
φ − φ − − − ω − ω − − ωω = φ + φ + φ
− − ω − ω+φ + φ +
∑ ∑ ∑ ∑
∑ ∑
25
Smoothing difference
difference equation:
=Cq b
Tx y xx xy yy( , , , , )= φ φ φ φ φq
O( )δ
26
Three-point stencil
(0) 0,Φ = (1) 1,Φ = *( )Φ δ = φ
δ
φx
−60
−40
−20
0
20
PSSD and exact
δ
δ
-1.0 -0.8 -0.6 -0.4 -0.2
φx
−20
0
20
40
60
δ
-1.0 -0.8 -0.6 -0.4 -0.2 0.0
φ*=2.
φ*=-0.5
φ*=0.5
φ*=-2. δ
φxx
−20
0
20
40
60
PSSD and exact
δ
δ
-1.0 -0.8 -0.6 -0.4 -0.2
φxx
−60
−40
−20
0
20
δ
-1.0 -0.8 -0.6 -0.4 -0.2 0.0
φ*=2.
φ*=-0.5
φ*=0.5
φ*=-2.
Gradient Laplacian
27
Analysis of Laplacian operator
One-dimensional case i
C
(i 1) ( 0.5)xn
− + α ξ −=
a=0.3: (a) nC=8; (b) nC=16.
28
Analysis of Laplacian operator
29
Analysis of Laplacian operator
Two-dimensional case
(a)particle; (b)exact (c)PS; (d)SD.
30
Analysis of Laplacian operator
31
Pure conduction
32
Pure conduction
33
Cavity flow
time0 20 40 60 80 100 120
ψmin
PS (nC=40)SD (nC=40)Ghia et al.
Re=100
Re=400
Re=1000
-0.05
-0.10
-0.10
-0.10
-0.15
34
Cavity flow
35
Cavity flow
PS (t 100)= SD (t 100)= FV
36
Cavity flow
PS (t 100 105)= − SD (t 100 105)= − FV
37
Cavity flow
38
Backward facing step flow
39
Backward facing step flow
40
Backward facing step flow
41
Backward facing step flow
42
Backward facing step flow
43
Backward facing step flow
44
CPU time
45
Axisymmetric: pure convection
r1.3 1.4 1.5 1.6 1.7
φ
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
exactnC=8nC=16nC=32nC=64
46
Axisymmetric: developing pipe flow
47
Axisymmetric: developing pipe flow
u/um
r/r0
0.0
0.2
0.4
0.6
0.8
1.0
Re=200Re=300Re=400Sparrow et al.
0 0.5 1
x*=0.001
x*=0.005
x*=0.02
x*=0.03
x*=0.05
x*=0.08
x*=0.12
48
Axisymmetric flow: confined impinging jet
x0.0 0.2 0.4 0.6 0.8 1.0
u
0.0
0.5
1.0
1.5
2.0
Chatterjee and Deviprasathpresent
49
Axisymmetric flow: sudden expansion
time0 20 40 60 80 100
Lr
0
2
4
6
8
10
Re=20.6
55.477.6
109
156.1
211.1
Re0 50 100 150 200 250
Lr
0
2
4
6
8
10
12
Hammad et al. (exp.)present (∆r=0.05)
50
Axisymmetric flow: sudden expansion
x0 2 4 6 8 10 12 14 16
uc
0.0
0.5
1.0
1.5
2.0
2.5
Hammad et al. (exp.)present (∆r=0.05)present (∆r=0.025)
Re=20.6 156.1
211.1
55.4 77.6109
time0 20 40 60 80 100
Lr
7
8
9
10
11
∆r=0.05
∆r=0.025
Hammad et al. (upper, exp.)
Hammad et al. (lower, exp.)
51
CPM
Characteristic particle method (CPM) – Hyperbolic Equation Systems
52
緣起 Water hammer project Godunov-type finite-volume
method Particle method
53
Governing Equations (1)
2H H a VV 0X g Xτ
∂ ∂ ∂+ + =
∂ ∂ ∂
V V HV g J 0X Xτ
∂ ∂ ∂+ + + =
∂ ∂ ∂
K /a1 ( K / E )( D / e )
=+
ρ = =f V
J V V2D
Λ
54
Governing Equations (2)
h h vv 0t x x
∂ ∂ ∂+ + =
∂ ∂ ∂
v v hv vt x x
λ∂ ∂ ∂+ + = −
∂ ∂ ∂
55
Governing Equations (3)
t x∂ ∂
+ =∂ ∂V VA S
hv
=
V
v 11 v
=
A
0vλ
= −
S
56
Characteristic Analyses(1)
−= 1A RΛR
= −
1 11 1
R+
= −
v 1 00 v 1
Λ
− − −∂ ∂+ =
∂ ∂1 1 1
t xV VR ΛR R S
+−
−
+ = = = −
1 ( h v ) / 2w( h v ) / 2w
W R V
h w wv w w
+ −
+ −
+ = = = −
V RW
57
Characteristic Analyses(2)
−∂ ∂+ =
∂ ∂1
t xW WΛ R S
w w( v 1) ( w w ) / 2t x
λ+ +
+ −∂ ∂+ + = − −
∂ ∂
w w( v 1) ( w w ) / 2t x
λ− −
+ −∂ ∂+ − = −
∂ ∂
1 v / 2 ( w w ) / 2v / 2 ( w w ) / 2
λ λλ λ
+ −−
+ −
− − − = = −
R S
58
Characteristic Analyses(3)
= +dx v 1dt
dw w wdt 2 2
λ λ++ −+ =
= −dx v 1dt
dw w wdt 2 2
λ λ−− ++ =
(i) Along the characteristic line of
,
(ii) Along the characteristic line of ,
59
特徵線法MOC
60
Particle Method Moving along characteristic line No convection term error Adaptive Dual particles Particle interactions
61
Regular Particles (Transmitted) 1. For the right-running characteristic,
2. For the left-running characteristic,
n 1 n np p p
x x ( v 1) t+ + ++ = + + ∆
,n 1 ,n ,n t / 2 ,np p p p
w ( w w )e wλ∆+ + + +
+ + + − − −= − +
n 1 n np p p
x x ( v 1) t− − −+ = + − ∆
,n 1 ,n ,n t / 2 ,np p p p
w ( w w )e wλ∆− − − −
− + − + − += − +
62
Regular Particles (Interpolated) (i) For the right-running characteristic,
(ii) For the left-running characteristic,
n 1 n 1 n 1 n 1R p p L,n 1 ,n 1 ,n 1n 1 n 1 n 1 n 1p L RR L R L
x x x xw w w
x x x x− + + −
+ − −
− − − −
+ + + +
− + − + − ++ + + +
− −= +
− −
n 1 n 1 n 1 n 1R p p L,n 1 ,n 1 ,n 1n 1 n 1 n 1 n 1p L RR L R L
x x x xw w w
x x x x+ − − +
− + +
+ + + +
+ + + +
+ + + + + ++ + + +
− −= +
− −
63
Regular Particles
64
Particle Management (i) Addition,
(ii) Merge,
65
Shock Particle (1) (i) Conservation form
(ii) Rankine-Hugoniot relation
het x
∂ ∂+ =
∂ ∂U F S
h
h
ee v
=
U
h
h 2 h
e ve v e
=
+ F
hL R R L( ) x [ ( ) ( )] t e x tδ δ δ δ− + − =U U F U F U S
L R L R( ) ( ) ( )σ − = −U U F U F U
x / tσ δ δ=
66
Removal of a particle
67
Boundary Condition
+ + + + ++
+
− + −=
−2 1 1 2
1 2
n 1 n 1 * n 1B L L L B L
B * n 1L L
( x x )w ( x x )ww
( x x )
68
Particle Trajectory
69
Linear Problem
x0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8
u
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
exactMOCparticle
∂ ∂+ =
∂ ∂u ua 0t x
( , ) ( , )= − ∆ − ∆u x t u x a t t t
70
Burger's Equation (1)
x-1.0 -0.5 0.0 0.5 1.0
u
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
exactt=0t=0.2t=0.6t=1.0
x-1.0 -0.5 0.0 0.5 1.0
u
0.25
0.50
0.75
1.00
1.25
exactt=0t=0.2t=0.6t=1.0
( , )u x 0 x= − . .( , )
. .1 0 x 0 5
u x 00 5 x 0 5
< −= > −
71
Burger's Equation (2)
x-1.0 -0.5 0.0 0.5 1.0
u
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5exactt=0t=0.1t=0.3t=0.5
time0.0 0.2 0.4 0.6 0.8 1.0
u
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
exactparticle method
u( x,0 ) sin( 2 x )= π
72
Sudden Valve Closure (1)
0h( x,0 ) h (1 x )= − 0v( x,0 ) v=IC:
0h(0,t ) h= 00
0
t tvv(1,t )
t t0<
= ≥BC:
73
Sudden Valve Closure (2)
−= 30v 10
74
Sudden Valve Closure (3)
x
h/h0
-202468
1012
presentanalytical (200 terms)
t=0.6t=1.0 t=0.8
x0.0 0.2 0.4 0.6 0.8 1.0
v/v0
-0.20.00.20.40.60.81.01.2
presentanalytical (200 terms)
t=0.6t=1.0 t=0.8
t=0.2
t=0.2
t
h/h0
-10
-5
0
5
10
t0 1 2 3 4 5 6
v/v0
-1.0
-0.5
0.0
0.5
1.0
x=0.1x=0.3x=0.5x=0.7analytical (200 terms)
75
Valve Closure Over Finite Time (1)
IC:
BC:
= Lh( x,0 ) h x 0v( x,0 ) v=
≤ −= + < < +
≥ +
00
0 00 0 C
C0 C
v for t tv ( t t )v(0,t ) [1 cos ] for t t t t2 t
for t t t0
π
76
Valve Closure Over Finite Time (2)
t
h/hL
-10
-5
0
5
10
t0 1 2 3 4 5 6
v/v0
-1.5
-1.0
-0.5
0.0
0.5
1.0
presentx=0.1 (analytical, 200 terms)x=0.3 (analytical, 200 terms)x=0.7 (analytical, 200 terms)x=0.9 (analytical, 200 terms)
t
h/hL
-5
0
5
10
t0 1 2 3 4 5 6
v/v0
-1.5
-1.0
-0.5
0.0
0.5
presentx=0.1 (analytical, 200 terms)x=0.3 (analytical, 200 terms)x=0.7 (analytical, 200 terms)x=0.9 (analytical, 200 terms)
t
h/hL
0
2
4
6
t0 1 2 3 4 5 6
v/v0
-1.0
-0.5
0.0
presentx=0.1 (analytical, 200 terms)x=0.3 (analytical, 200 terms)x=0.7 (analytical, 200 terms)x=0.9 (analytical, 200 terms)
tC=1.0
tC=5.0
tC=3.0
77
Start-up and Evolution to Steady State(1)
IC:
BC:
0h( x,0 ) h= v( x,0 ) 0=
0h(0,t ) h= 00
0
t thh(1,t )
t t0<
= ≥
78
Start-up and Evolution to Steady State(2)
t
h/h0
0.00.20.40.60.81.01.2
t0 2 4 6 8 10
v/v0
-0.2
0.0
0.2
0.4
0.6
0.8presentx=0.3 (analytical, 200 terms)x=0.7 (analytical, 200 terms)
79
Periodic Forcing (1)
0h( x,0 ) h (1 x )= − 0v( x,0 ) v=IC:
BC: 0 0h(0,t ) h a sin( t )ω= + h(1,t ) 0=
80
Periodic Forcing (2)
t
h/h0
0.2
0.4
0.6
0.8
1.0
1.2
t0 4 8 12 16 20 24
v/v0
0.98
1.00
1.02
1.04
presentx=0.1 (analytical, 200 terms)x=0.3 (analytical, 200 terms)x=0.5 (analytical, 200 terms)x=0.7 (analytical, 200 terms)
81
Sound speed variations (1)
x0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
u/u0
-0.20.00.20.40.60.81.01.2
h/h0
02468
101214
present, ∆x=1/400MOC, ∆x=1/800MOC, ∆x=1/3200
t=0.6t=1.0 t=0.75
t=0.6t=1.0 t=0.9
t=0.9t=1.1
t=1.1 t=0.75
t=1.1
t=1.1
82
Sound speed variations (2)
83
Sound speed variations (3)
x0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
u/u0
-0.20.00.20.40.60.81.01.2
h/h0
02468
101214
a*=0.8a*=1.0a*=1.2
t=0.85t=1.2
t=1.0
t=0.85
t=1.2
t=1.0
t=1.2
t=1.2
84
Branch
x0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
h/h0
t=0.5t=0.8, a*=1.0t=1.0, a*=1.0t=1.3, a*=1.0t=0.8, a*=1.2t=1.0, a*=1.2t=1.3, a*=1.2t=0.8, a*=0.8t=1.0, a*=0.8t=1.3, a*=0.8
t=0.8
t=1.3
t=1.0
Branch B
Branch A
t=1.3
t=1.3
t=1.0
t=1.3
t=1.3
t=1.0
t=0.8
t=0.8
0
5
10
5
10
15
15
Branch LBranch R
RJIL
85
Sudden Valve Closure (1)
0h( x,0 ) h (1 x )= − 0v( x,0 ) v=IC:
0h(0,t ) h= 00
0
t tvv(1,t )
t t0<
= ≥BC:
86
Sudden Valve Closure (2)
x
h
-0.1
0.0
0.1t=0.5t=1.0t=1.5 t=2.0
t=4.0 t=2.5t=3.0t=3.5
x0.0 0.2 0.4 0.6 0.8 1.0
v
-0.1
0.0
0.1
t=0.5t=1.0
t=1.5 t=2.0 t=2.5t=3.0
t=3.5 t=4.0
−= 10v 10
87
Shock Tube Problem (1)
x0.0 0.2 0.4 0.6 0.8 1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
exacth (t=0.2)h (t=0.5)v (t=0.2)v (t=0.5)
=L L( h , v ) ( 2,0 ) =R R( h , v ) (1,0 )
88
Shock Tube Problem (2)
=L L( h , v ) (1,0 ) = −R R( h , v ) (1, 1)
x0.0 0.2 0.4 0.6 0.8 1.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
exacth (t=0.2)h (t=0.5)v (t=0.2)v (t=0.5)
89
Shock Tube Problem (3)
= −L L( h , v ) (1, 1) =R R( h , v ) (1,0 )
x0.0 0.2 0.4 0.6 0.8 1.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
exacth (t=0.2)h (t=0.5)v (t=0.2)v (t=0.5)
90
Valve Action (1)
p
1.00
1.01
1.02
x
p
1.0
1.1
1.2
x0 2 4 6 8 10
p
0.5
1.0
1.5
2.0finite volume (UD)finite volume (van Leer)particle method
∆u=∆p=0.01
∆u=∆p=0.1
∆u=∆p=0.5
91
Valve Action (2)
x0 2 4 6 8 10
p
0.5
1.0
1.5
2.0
x
p
1.0
1.5
2.0
p
1.0
1.5
2.0finite volume (UD)finite volume (van Leer)particle method
t=14
t=8
t=2
92
Valve Action (3)
x
time
0 2 4 6 8 100
2
4
6
8
10
12
14
x
time
0 2 4 6 8 100
2
4
6
8
10
12
14
x
time
0 2 4 6 8 100
2
4
6
8
10
12
14
UD Van Leer Present
93
Open Channel
x
0.2 0.3 0.4 0.5 0.6 0.7 0.8
v
0.0
0.5
1.0
1.5
2.0 exactSPH (n=1000), Chang et al.SPH (n=5000), Chang et al.present (n=64)
x
0.2 0.4 0.6 0.8 1.0
v
0.0
0.5
1.0
1.5
2.0 exactFV (n=120), Ying et al.present (n=64)
94
Tidal bore
x0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
b
-5
-4
-3
-2
-1
0
1
2
3
4
5
α+β tanh[γ(x-0.7)]
x0.0 0.2 0.4 0.6 0.8 1.0
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
t=0.2t=0.4t=0.6t=0.8t=1.0
η
v
x
t
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.05
1.45
1.075
1.1
1.025
1.5
1.35
1.5
1.45
1.4
1.3
1.25
1.2
1.15
95
Two-phase flow
x0.0 0.2 0.4 0.6 0.8 1.0
p/pref
0.0
0.2
0.4
0.6
0.8
1.0
1.2
v
0.0000
0.0002
0.0004
0.0006
0.0008
0.0010
0.0012
exactp/pref
v
/ =L Rp p 100
96
Traffic flow (LWR model)
x-1.0 -0.5 0.0 0.5 1.0
ρ
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
exactt=0.5
1.02.0
3.0
t=0.
0.25
= − −L R L( x,0 ) sin[( x x ) /( x x )]ρ π
t0.0 0.5 1.0 1.5 2.0 2.5 3.0
xsh
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
exactpresent
0.3 0.4 0.5 0.6 0.7-0.64
-0.63
-0.62
-0.61exactδx=0.05δx=0.025
97
Traffic flow (Payne model): Branch
98
Traffic flow (Payne model): Red-Green
t
0.15
0.20
0.25ρu
t0 1 2 3 4 5 6 7 8 9 10
0.10
0.15
0.20
0.25
x=0.3
x=0.7
0.15
0.16
0.16
0.16
0.16
0.16
0.17
0.17
0.17
0.17
0.18
0.18
0.18
0.18
0.18
0.18
0.18
0.19
0.19
0.19
0.19
0.19
0.19
0.19
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.21
0.21
0.21
0.21
0.21
021
0.21
0.21
0.21 0.22
0.22 0.22
0.22
0.22
0.22
0.22
0.23
0.230.23
0.23
0.23
0.23
0.23
0.24
0.24
0.24
0.240.24
0.24
0.25
0.25
0.25
0.25
0.26
0.26
0.260.27
0.27
0.270.27
0.27
0.280.28
0.29
x
t
0 0.2 0.4 0.6 0.8 19
9.2
9.4
9.6
9.8
10
99
Gas dynamics
x0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0.0
0.2
0.4
0.6
0.8
1.0
1.2
exactUDMinModESSGpresent
density
pressure
velocity
100
Gas dynamics
x0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
ρ
0.0
0.2
0.4
0.6
0.8
1.0
1.2
exactUDMinModESSG
0.6 0.7 0.8 0.90.1
0.2
0.3
0.4
0.5present