[PPT] - Nonlinear 3D FE-Stability analysis of suction pile Ralf Lampert 1 , PowerPoint Presentation

SLIDE 1

SLIDE 2

Nonlinear 3D FE-Stability analysis of suction pile

Ralf Lampert1, Stefan Eckardt2, Roger Schlegel2, Kjetil Rognlien3, Frode Halvorsen3

1) Dynardo US, Inc. San Francisco, USA; 2) Dynardo GmbH Weimar, Germany; 3) EDRMedeso AS, Norway *Acknowledgments to Project Partner: TechnipFMC plc , DNV GL

SLIDE 3

Nonlinear 3D FE-Stability analysis of suction pile 3

Founded: 2001
More than 60 employees,
ffices at Weimar, Vienna and

San Francisco

Leading technology companies Daimler,

Bosch, E.ON, Nokia, Siemens, EADS are supported

Software Development

Dynardo is engineering specialist for CAE-based sensitivity analysis, optimization, robustness evaluation and robust design optimization

project work
support
know how transfer / seminar
embedded engineer
workflow and process automation

CAE-Consulting Dynardo – dynamic software & engineering GmbH

SLIDE 4

Nonlinear 3D FE-Stability analysis of suction pile 4

Task - Stability analysis of suction pile construction

Suction piles are anchoring structures

in the seabed.

They are used e.g. for foundation

construction for oil & gas applications

r wind turbines.
For installation, an internal

underpressure is created inside the pile, so that the pipe is pulled into the seabed.

For proof of stability of the suction pile

the soil-structure-interaction is important.

The finite element method has

especially proven to be suitable for the simulation and verification of suction pile constructions.

SLIDE 5

Nonlinear 3D FE-Stability analysis of suction pile 5

Workflow - 3D FE-Stability analysis of suction piles

Steps for Solution with ANSYS optiSLang:

1. 3D ANSYS model build up for the suction pile and soil
2. Definition of the model & soil parameter,

Definition of the nonlinear load history (primary stress state, pore pressure state, suction procedure)

3. Introduction of imperfection (from Eigenvalue Buckling Analysis)
4. Non-linear stability analysis considering structural imperfections and

soil-structure interaction with soil (Mohr-Coulomb) material models for different load cases

5. Model validation & Sensitivity analysis for checking model quality and

uncertainty of boundary conditions

6. Proof of serviceability and stability

SLIDE 6

Nonlinear 3D FE-Stability analysis of suction pile 6

Calculation Flow in ANSYS optiSLang

SLIDE 7

Nonlinear 3D FE-Stability analysis of suction pile 7

Finite Element Model Properties

Geometry, Mesh

prefered brick mesh, 0.5 – 1 Mio DOF suction pile & soil

deformation boundary conditions perpendicular to the surface on all

sides

on the bottom side soil layer with uv = 0 on bottom surface
on the top side without coupling between top plate and inner soil

boundary conditions

uvertical = 0

n Soil Layers: based on geotechnical report

SLIDE 8

Nonlinear 3D FE-Stability analysis of suction pile 8

σN RES Φ C ft

tension shear material law for joints shear and tensile failure tension shear material law for rock/concrete shear and tensile failure is a library of elasto-plastic material models, which describe the behavior of natural (rock, soil, wood) and of artificial (steel, concrete, masonry) materials for civil engineering and geotechnical applications Unique features:

consistent elasto-plastic algorithms allowing the efficient numerical handling of multiple failure surfaces within the

framework of multi-surface plasticity

powerful combination of different yield conditions/failure surfaces
very realistic nonlinear material models which are based on engineering material parameters with a clear physical

meaning: e.g. friction angle, dilatancy angle, cohesion, tensile strength, compressive strength

identification and visualization of failure mechanisms based on plastic activities

nonlinear hardening and softening behavior

ANSYS Geomechanical Toolbox

Material modeling

SLIDE 9

Nonlinear 3D FE-Stability analysis of suction pile 9

Soil structure interaction

Modeling nonlinear soil behaviour with multisurface Mohr

Coulomb / Rankine material models

Modeling contact joint between soil and suction pile with

nonlinear friction contact elements or anisotropic Mohr Coulomb friction model with tension cut off aproach

t RES

s N

tan

Re

     c F

N s

  

1

  

t N

f F 

SLIDE 10

Nonlinear 3D FE-Stability analysis of suction pile 10

Model Check: Comparison Earthpressure

SX horizontal stresses compared with analytical calculation (TIME=1) Left: SX (global coordinates) Right: Pathplot (SX)

pathplot Layer 1: f1, r1 Layer 2: f2, r2 Layer 3: f3, r3

SLIDE 11

Nonlinear 3D FE-Stability analysis of suction pile 11

Model Check: physical plausibility for nonlinear load history

ux horizontal displacement of complete soil (cylindrical coordinate system)

Loadstep1: Gravity (t=1) Loadstep2: (suction pressure) (t=2)

SLIDE 12

Nonlinear 3D FE-Stability analysis of suction pile 12

Model Check: physical plausibility for nonlinear load history

Three positions with two elements each – one from the inside and one form the outside Left: radial stress SX right: radial displacement UX

G I K In (core) A C E Out (soil)

x(analytical)=k0·r·g·z=~8000Pa x(analytical)= ~3800Pa

SLIDE 13

Nonlinear 3D FE-Stability analysis of suction pile 13

Model Check: parameter studies for inner soil

For conservative calculation we reduced the strength (inner friction angle) and stiffness (G, E) of the inner soil (to avoid unrealistic „container effect“ for inner soil). Although a range of 30-35° for friction angle F in geotechnical report is given, we investigated a reduction up to a F of 0°. This decrease is a result of the inner suction pressure, which will generate a perfusion of the soil and therefore loosen the soil . This leads to following parameters and their ranges for a given cohesion:

friction angle Fcore (inner soil):

0-30°

density rcore (inner soil):

0-50% reduction

young’s modulus Ecore (inner soil):

values corresponding to F

poisson ration ncore (inner soil):

values corresponding to F

SLIDE 14

Nonlinear 3D FE-Stability analysis of suction pile 14

Model Check: parameter studies for inner soil

1) considering a

chosen imperfection

Load vs. ux displacement (cylindrical coordinate system):

Overview of some Calculations: Load-Displacement (ux – radial displacement) This figure shows the ux radial displacements of the suction pile at the point of its largest deflection. Blue line depicts the reference calculation (nonlinear spring model) reducing r reducing f, n (additionally) Point of Failure start design without reduction of inner soil

SLIDE 15

Nonlinear 3D FE-Stability analysis of suction pile 15

Mesh sensitivity: on the DNV GL simplified plane strain model test

𝜏𝑜𝑝𝑠𝑛𝑏𝑚 = 5.14 𝑡𝑣

SLIDE 16

Nonlinear 3D FE-Stability analysis of suction pile 16

Mesh sensitivity: on the DNVGL simplified plane strain model test

skirt

SLIDE 17

Nonlinear 3D FE-Stability analysis of suction pile 17

Mesh sensitivity: on the DNVGL simplified plane strain model test

𝜏𝑜𝑝𝑠𝑛𝑏𝑚 = 5.14 𝑡𝑣

total displacements USUM (m) plastic strains EPPLEQV volumetric elastic strains EPEL VOL

SLIDE 18

Nonlinear 3D FE-Stability analysis of suction pile 18

Practical project application – Johan Sverdrup oil field

A quarter-model of the a full ITS (Integrated Template Structure) assembly used in the analyses. The surrounding soil is modelled with a width of 14x14m and a depth of 2m below the suction pile’s tip.

SLIDE 19

Nonlinear 3D FE-Stability analysis of suction pile 19

Practical project application – Johan Sverdrup oil field

Buckling mode von Mises Stresses radial displacement equivalent plastic strain

Results at structural collapse

SLIDE 20

Nonlinear 3D FE-Stability analysis of suction pile 20

Practical project application – Johan Sverdrup oil field

pressure displacement (m)

0,0 0,5 1,0 1,5 2,0 2,5 3,0 Soil Spring Stifness – Layer 1

Old Spring – Oval NewSpring – 6 Waves New – Springs for 6 waves – Mod 1.5 New – Springs for 6 waves – Mod 2 Continuum Model

162% 100%

SLIDE 21

Nonlinear 3D FE-Stability analysis of suction pile 21

ANSYS optiSLang – variation analysis

SLIDE 22

Nonlinear 3D FE-Stability analysis of suction pile 22

is a general purpose tool for variation analysis

using CAE-based design sets and/or data sets for the purpose of

sensitivity analysis
design/data exploration
calibration of virtual models to tests
optimization of product performance
quantification of product robustness and reliability
Robust Design Optimization and Design for Six Sigma

serves arbitrary CAX tools with support of process integration process automation workflow generation

Excellence of

SLIDE 23

Nonlinear 3D FE-Stability analysis of suction pile 23

Design Improvement

Optimize design performance

Design Quality

Ensure design robustness and reliability

Design Quality

Ensure design robustness and reliability

Model Calibration

Identify important model parameter for the best fit between simulation and measurement

Model Calibration

Identify important model parameter for the best fit between simulation and measurement

Design Understanding

Investigate parameter sensitivities, reduce complexity and generate best possible meta models

Design Understanding

Investigate parameter sensitivities, reduce complexity and generate best possible meta models

CAE-Data Measurement Data Robust Design

Design Improvement

Optimize design performance

RDO – Robust Design Optimization

SLIDE 24

Nonlinear 3D FE-Stability analysis of suction pile 24

ANSYS optiSLang – sensitivity analysis & optimization

4. Show and explain the best design
1. Start with a sensitivity study using the LHS

Sampling

3. Define the objective function

Run an ARSM, gradient based or biological based optimization Algorithms Understand the Problem using CoP/MoP Search for Optima Scan the whole Design Space

2. Identify the important parameters and responses,

understand the problem, reduce the problem

SLIDE 25

Nonlinear 3D FE-Stability analysis of suction pile 25

ANSYS optiSLang – model calibration / parameter identification

1) FE-modeling / model validation

check model size,
check model plausibility with reference analyses
check describtion of the basic physical

phenomenon

2) Sensitivity Analysis – Scan the Design Space (LHS)

identify sensitive parameters and

responses

check the variation responses

versus measurements

reduce parameter space & extract

start values for optimization

R3 R2 R1

3) Optimization – find the best fit

define the right objective function
choose an optimizer depending on the sensitive
ptimization parameter dimension/type

MOP

SLIDE 26

Nonlinear 3D FE-Stability analysis of suction pile 26

ANSYS optiSLang – robustness & reliability analysis

1. Define the robustness space using

scatter range, distribution and correlation

3. Check the variation interval
4. Build MOP

, identify the most important scattering variables

5. Reliability Analysis

Define Limite state function, choose right algorithms

2. Scan the robustness space by producing

and evaluating n (100) Designs

Robustness Analysis

SLIDE 27

Nonlinear 3D FE-Stability analysis of suction pile 27

Summary

Design and optimization of suction pile construction with ANSYS optiSLang:

1. Build the parametrized 3D ANSYS model for the suction pile and soil
2. Definition of the model & soil parameter,

Definition of the nonlinear load history (primary stress state, pore pressure state, suction procedure)

3. Introduction of imperfection (from Eigenvalue Buckling Analysis)
4. Non-linear stability analysis considering structural imperfections and soil-

structure interaction with soil (Mohr-Coulomb) material models for different load cases

5. Sensitivity analysis for checking model quality and uncertainty of boundary

conditions

6. Optimization of the suction pile construction for minimal costs with sufficient

stability

7. Proof of serviceability and robustness

SLIDE 28

Nonlinear 3D FE-Stability analysis of suction pile 28

Thank you for your attention.

For more information please visit our homepage www.dynardo.de