numerical simulation of - emshipm120.emship.eu/documents/masterthesis/2017/tsai ssu.pdfnumerical...
TRANSCRIPT
Numerical simulation of surface ship hull beam whipping response due to submitted underwater explosion Presenter / Ssu-Chieh Tsai
Supervisor / Pr. Hervé Le Sourne 1 March 2017, Rostock
1 NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Motivation
UNDEX Plume Above-Surface Effects
2 NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Background
3
• Complete bubble migration process
• First shock wave (Mauricio,2015) o Exponential decay o Empirical approach o Very short time duration (ms)
• Bubble oscillation phase o Non-linear o Longer time duration o Motion of bubble migration o Radius and vertical displacement
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Methodology – 3 models
4
• Waveless model
- Consider as an ideal fluid - Without damping effect and the effect of gas bubble pressure inside bubble
• DAA model (Doubly Asymptotic Approximation)
- Considering the interaction between the liquid and gas bubble
- Including damping effect in the liquid
• Empirical model
- Peak approximation method
- Restricted to experienced coefficients, specific material cases
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Development of Matlab programs • Developed code in Matlab
• Empirical and analytical models
• Calculate motion of bubble
• Pressure loads
• Test program by 3 charge cases
5
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Comparison on 3 different UNDEX cases • The 3 models are tested on:
Case 1: Barras Guillaume’s Case
Case 2: Hunter & Geers’ Case
Case 3: Keith G, Webster’s Case
Case 1 Description mc TNT charge mass, mc= 500 kg
Distance from charge to free surface, = 50 m r Distance from charge to standoff point, r = 50 m
Density of charge, = 1600 kg/m3
Case 2 Description mc TNT charge mass, mc= 0.3 kg
Distance from charge to free surface, = 92 m Density of charge, = 1500 kg/m3
Case 3 Description mc TNT charge mass, mc= 1.45 kg
Distance from charge to free surface, = 178 m Density of charge , = 1500 kg/m3 Radial distance from charge,
6
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
• Comparison with experience (Case 1) Experience: Barras Guillaume, Numerical simulation of underwater explosions using an ALE method. The pulsating bubble phenomena. (2012)
7
Case 1 / Pressure Case 1 / Radius: a
DAA model is close to Experience model !
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
• Comparison 3 models for Case 2 & Case 3
8
↓ With damping
↓ With damping
Case 2 / Radius: a
Case 3 / Radius: a
↓Without damping
↓Without damping Case 3 / Pressure
Case 2 / Pressure
← not realistic
← not realistic
Conclusions
Analytical models can be used for various charge materials, mass and water depth
DAA model is more representative of the reality than waveless model
9 NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Example 1 : Clamped plate model
Clamped B.C. ( Tx, Ty, Tz, Rx, Ry, Rz) Symmetrical B.C. (Tx, Ry, Rz)
Girder Girder
Frame
Frame Symmetrical B.C. (Ty, Rx, Rz)
4.9 m
8.4 m
10
[30 mm]
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
• Case1: mc= 500 kg , 𝑑𝑑𝑖𝑖 = 50 m, 𝜌𝜌𝑐𝑐= 1600 kg/m3
Results of clamped plate
11
The deformation seems to be realistic !
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
• Charge Case 1: mc= 500 kg , 𝑑𝑑𝑖𝑖 = 50 m, 𝜌𝜌𝑐𝑐= 1600 kg/m3
• Length: 150 m, Breadth: 20m, Draft: 8 m
Example 2 : Semi-cylinder model
12
[10 mm]
[20 mm]
[80 mm]
½ L ¼ L
B1
B2
B3
D1 D2
D3
S1
S2
S3
50 m
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Pressure loads for semi-cylinder
13
• Charge Case 1: mc= 500 kg , 𝑑𝑑𝑖𝑖 = 50 m, 𝜌𝜌𝑐𝑐= 1600 kg/m3
B1 B2
B3 S1
S2
S3
14
• Results of semi-cylinder model
0.0E+00
1.0E+07
2.0E+07
3.0E+07
4.0E+07
5.0E+07
6.0E+07
7.0E+07
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Ener
gy (J
)
Time (s)
Energies in LS-DYNA
Kinetic Energy Internal Energy Total Energy
Hourglass Energy External Work
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Z-D
ispl
acem
ent (
m)
Time (s)
Z-displacement on Bottom (LS-DYNA)
B1_Elm_9906 B2_Elm_10020 B3_Elm_9681
-0.20
-0.10
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0 0.2 0.4 0.6 0.8 1 1.2Nod
al D
ispl
acee
nt (m
) Time (s)
Difference of Z-Displacement at Bottom of Cylinder (LS-DYNA)
UZ (B1-B3)
↓High energy ↓High energy
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
The deflection is realistic.
Analysis process for ANSYS
Model Preparation
• Transient solution • Calculate for all model • Time consumption
Pressure Calculation Direct Integration Method
Modal Superposition
Constraint DOF to all nodes on hull
Static analysis for reaction force
Modal analysis
Transient solution
Create modal basis
Export hull model
Reaction force calculation to nodes
Fluid field mesh
Structural model
OR
15 NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
• Objective: Convert code from MATLAB to ANSYS language
Example 2: Semi-cylinder model
16 NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Pressure load calculation is validated in ANSYS program !
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Results of semi-cylinder model
17
• Static analysis
• Transient solution by model superposition
o Displacement is should be zero at time 0
o Too high deflection > 14m !!!
B1 B2 B3
• Modal analysis
Deflection is not realistic !
Example 3 : Frigate ship model
18 NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Charge location Charge location
64.7
5 m
40
m
Case Description SF Shock factor = 0.6 mc TNT charge mass, mc= 1296 kg
Distance from charge to free surface, = 64.75 m r Distance from charge to standoff point, r = 60 m
Density of charge, = 1600 kg/m3
• Length of all: 95.0 m • Breadth: 14.0 m • Draught: 4.75 m
-0.20
-0.15
-0.10
-0.05
0.00
0.05
0.10
0.15
0.20
0 0.2 0.4 0.6 0.8 1 1.2
Nod
al D
ispl
acee
nt (m
)
Time (s)
Difference of Z-displacement at Bottom (Frigate ship)
UZ(B1-B3)
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Results of Example 3
19
• Transient solution by modal superposition
Deflection seems to be more realistic
Displacement should be zero at time 0
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Conclusion
20
• DAA analytical model, representative of the reality, has been chosen
• Secondary bubble oscillation phase has significant influence on structure
• Model superposition method leads to unrealistic results
This method is suitable only for small displacements (it is not the case here)
• Direct integration method
NUMERICAL SIMULATION OF SURFACE SHIP HULL BEAM WHIPPING RESPONSE DUE TO SUBMITTED TO UNDERWATER EXPLOSION
Future work
21
• Perform shock response analysis for one or several embarked materials using
Dynamic Design Analysis Method (DDAM)
• Examples : Shaft line or / and pipes line