Advanced Design Methods for Offshore Wind …...May 2016 Final report approved After May 2016...

Advanced Design Methods for Offshore Wind Foundations

Professor Byron Byrne Ørsted / Royal Academy of Engineering Research Chair

University of Oxford

Danish Geotechnical Society

7 February 2019

Structural Options

(g)(f)(e)(d)(c)(b)(a)

Images from various websites

Danish Geotechnical Society 7 February 2019

Contact: [email protected] Copyright University of Oxford

Univer

sity o

f Oxfo

rd

http://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=0ahUKEwieyrzll87XAhXEBcAKHR1VCHIQjRwIBw&url=http://logos.wikia.com/wiki/%C3%98rsted&psig=AOvVaw0km2RM-7wEZs1VMm8K8goA&ust=1511302791652038

https://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&cad=rja&uact=8&ved=2ahUKEwjP89e80b7bAhWD1RQKHVh6Dz8QjRx6BAgBEAU&url=https://www.spilasers.com/news/new-royal-academy-engineering-research-chair/attachment/raeng-logo-2013/&psig=AOvVaw0aavuy-jmVyQa3xtne58aJ&ust=1528360674671294




Monopile: Monotonic Loading

H

e

D

L

H

vg

Ultimate Load Soil conditions at failure

Initial Loading Small strain soil behaviour Natural frequency is “tuned”

For Monopiles L/D = 2 to 6 e/D = 5 to 15

Photos from Dan Kallehave (Ørsted)

Industrialised Design

Pile length

Wal

l th

ickn

ess

Diameter

L?

t?

D?



Univer

sity o

f Oxfo

rd





Problem Definition

Pile displacement, y

Spring force, p

pu

yc

𝑝(𝑦) = 𝑝𝑢2

𝑦

𝑦𝑐

13

𝑓𝑜𝑟 𝑦 ≤ 8𝑦𝑐

𝑝𝑢 𝑓𝑜𝑟 𝑦 > 8𝑦𝑐

Simplified soil-structure

interaction models allow for the many

calculations needed to optimise wind farm foundations

PISA Project Overview – 2.5 years and £3.5m 3D FE Field Tests

Design

Validate

MG

HG

z v

p(z,v)

m(z,y)

HB(vB) MB(yB)

Simplified 1D

Model

Accurate Response Prediction

Apply

S

Mb

M

H

τ

τ

p

p

z

0

5

10

15

20

0 0.2 0.4 0.6 0.8 1

Ho

rizo

nta

l Lo

ad (

MN

)

Ground Level Displacement (m)

p-y method

3D FE

1D Model



Univer

sity o

f Oxfo

rd





PISA Project Structure and Timetable

Aug 2013 Project start date Nov 2013 Field test tender begins Aug 2014 Design methods Dec 2014 Pile installation Jan 2015 Pile testing start Jul 2015 Pile testing complete Nov 2015 Field test factual results Feb 2016 Close-out workshop May 2016 Final report approved After May 2016 Implementation

Numerical Modelling

Ho

rizo

nta

l Lo

ad

Ground Level Displacement

capacity: - soil conditions at failure

operational loads: - small strain behaviour

0

2

4

6

8

10

12

14

0 50 100 150 200 250 300

De

pth

z(m

)

su TXC (kPa)

TXC - historic data

HSV (PISA)

TXC 38mm (PISA)

TXC 100mm (PISA)

TXC 100mm (GEO)

Simulated

Sand Layer

0

2

4

6

8

10

12

14

0 5 10 15 20

De

pth

, z(m

)

qc (MPa)

Max

Mean

Min



Univer

sity o

f Oxfo

rd





Field Test Campaign – 28 Pile Tests D = 2.0 m

21 m12 m

D = 0.76 m

D = 0.27 m

z

Stick up inclinometer

Load cell

Displacement transducer

Fibre Optic SG

Inclinometer

Extensometer

h

D

L

Test Pile Reaction pile

HG

vG/D (%)0 2 4 6 8 10 12

UL/RL

98

7

65

4

3

2

10

Cowden Test Results: D = 2 m , L/D = 5.25

0

0.5

1

1.5

2

2.5

0 50 100 150 200

Ho

rizo

nta

l Lo

ad (

MN

)

Ground Level Displacement (mm)

0.1D

p-y method

3DFE

Field Test



Univer

sity o

f Oxfo

rd





PISA Design Model

MG

HG

Tower

Ground level

Mon

opile

Distributed lateral load

Vertical shear stresses at pile/

soil interface

Shear force and moment applied at the pile base.

MG

HG

Tower

Ground level

Distributed lateral load,

p(z,v)

Distributed moment m(z,y)

z

v

HB(vB)

MB(yB)

Timoshenko beam finite elements

3D Finite Element Modelling Extract Information

Normalise and Parameterise

k

yu

xu

n

y u 0

5

10

15

20

0 0.2 0.4 0.6 0.8 1

H (

MN

)

vG (m)

Apply to 1D Model

Design Case: L/D = 4, M/HD = 10 and D = 8.75m

Calibration (11 Calculations) D = 5, 7.5, 10 m

L/D = 2, 6 M/HD = 5, 15

t/D = 60, 80, 110 Design Improved from L/D = 6.2 to L/D = 4: SAVING = 35%!

0

2

4

6

8

10

12

14

16

0 2 4 6 8 10

M/H

D

L/D

Calibration cases

Design cases

0

5

10

15

20

25

30

0 0.2 0.4 0.6 0.8 1

Ho

rizo

nta

l Lo

ad (

MN

)

Ground Level Displacement (m)

0.1D

p-y: L/D = 6.2

p-y: L/D = 4

1D Model

3DFE



Univer

sity o

f Oxfo

rd





PISA Design Method Application

Lateral response prediction

Soil classification Basic strength and stiffness parameters from SI

Lookup table of parameters for given soil classification

Soil reaction curves

Pile geometry

Loads

Detailed strength and stiffness parameters from SI

Soil reaction extraction and parameterisation

Soil reaction curves

Array geometry

Loads

Lateral response prediction

Soil constitutive model calibration and FE analysis

Ru

le-B

ased

Nu

mer

ical

-Bas

ed

MG

HG

p

m

HB

MB

PISA Layered PISA2

Chalk

ALPACA

Foundation Scour

Cyclic Modelling

Monitoring

MoDeTo



Univer

sity o

f Oxfo

rd






https://www.google.co.uk/url?sa=i&rct=j&q=&esrc=s&source=images&cd=&ved=0ahUKEwjxnaKtpIDYAhUMtRQKHcLgBI8QjRwIBw&url=https://be.linkedin.com/jobs/view/electrical-package-engineer-at-parkwind-514488624&psig=AOvVaw2VjGsDZ_Vb_kKKZHiKBASS&ust=1513024156836520


PLAXIS Monopile Design Tool: MoDeTo

• Commercial application of PISA Method by PLAXIS

– Collaboration with Oxford University and Fugro

• Validated against experimental results from both test sites

• Rule-based and numerical-based design using user-defined or automatically calibrated soil response curves

– Rule-based: Stand-alone tool. User-defined SRC

– Numerical-based: SRC calibration from PLAXIS 3D FE model

• More info: http://www.plaxis.com/modeto

e

C)(A) (E/F)

dt

(B) (

d=L/2dt

(D)

e

(PC)

PISA Phase 2 Layered Soils

PISA Phase 2 London clay

Bothkennar clay

Dunkirk sand (relative density 45%, 60%, 90%

Calibrated 1D model

(a) Application to Layered Soil Homogeneous Soils

(b) General Dunkirk Sand Model

e

C)(A) (E/F)

dt

(B) (

d=L/2dt

(D)

e

(PC)

PISA Phase 1 Cowden till

Dunkirk sand

(relative density 75%)



Univer

sity o

f Oxfo

rd



http://www.plaxis.com/modeto



Example PISA Phase 2 Output Pile D2: D=8.75 m, L/D=4, h=87.5 m, t=91 mm

0

0.1

0.2

0.3

0.4

0.5

0 0.0004375 0.000875H

(M

N)

vG (m)

Small displacement

0

5

10

15

20

25

30

35

40

0 0.4375 0.875

H (

MN

)

vG (m)

Ultimate response

-35

-25

-15

-5

0 1750 3500

De

pth

(m

)

Bending Moment (MNm)

Hult/2Hult

020 0.005 0.01H

( M

vG (m)Small displacement3D FE 1D (parametric)

0

5

10

15

20

25

30

35

40

0.74

0.78

0.82

0.86

0.90

0.94

0.98

1.02

1.06

1.10

1.14

1.18

1.22

1.26

1.30

1.34

1.38

1.42

1.46

1.50

1.54

Fre

qu

en

cy

Load (1D Model) / Load (3D FE)

Layered

Homogeneous

Mean = 1.01 CoV = 4.7%

0

5

10

15

20

25

30

35

0.74

0.78

0.82

0.86

0.90

0.94

0.98

1.02

1.06

1.10

1.14

1.18

1.22

1.26

1.30

1.34

1.38

1.42

1.46

1.50

1.54

Fre

qu

en

cy

Load (1D Model) / Load (3D FE)

Layered

Homogeneous

Mean = 1.01 CoV = 7.2%

PISA Phase 2 Summary



Univer

sity o

f Oxfo

rd





• A number of conference papers available including keynote for the 2017 London SUT OSIG conference.

• An overview paper will be presented at 2019 OTC

• 8 Papers on the field testing / numerical modelling / design methods submitted for journal publication – 5 now accepted for publication in Géotechnique

– 3 under review

• Papers to come through in the near future include: – Papers on the cyclic loading field tests

– Papers reporting the PISA Phase 2 project

PISA Publications

Application to Suction Caissons (Suryasentana 2018)

• Similar Winkler framework as PISA (but 6DoF loading)

• OxCaisson: Family of fast design methods calibrated by 3D FE

– FLS analysis in linear elastic soil

– SLS analysis in non-linear elastic soil

– ULS analysis in elasto-plastic soil

Displacement

Load

FLS SLS ULS

PlasticNon-linear Elasticor

Elasto-Plastic

Linearelastic



Univer

sity o

f Oxfo

rd





Monopile Cyclic Loading: Basics

H

vg

H

vg

One Way Loading

Two Way Loading

• Monotonic loading response is an essential input for any cyclic loading calculation

• Cyclic loading models must capture hysteresis, ratcheting behaviour and stiffness change

• Ultimate capacity after cycling • Cyclic loading is irregular / pseudo random

Cyclic Testing



Univer

sity o

f Oxfo

rd





Cyclic Results from PISA – Current Design Approach (?)

Lo

ad

Displacement

Backbone ResponsePISA or p-y calculation

Factored Response for Cyclic Behaviour

• Hyperplasticity , an approach to plasticity theory based on thermodynamic principles, developed by Prof. Guy Houlsby and Prof. Sasha Puzrin

• Entire model specified by just two functions with output derived using standard mathematical procedures

• Can be applied at 0D macro model, 1D generalised Winkler model or at the 3D continuum level

Hyperplastic Accelerated Ratcheting Model (HARM)

E0

E1

E2

ENs

e

s

e1 e2 eNs

s1

s2

sNs

H0

s

e

k1

H1

kNs

HNs

aNs a1

R

ar

q

M

N cycles

10N cycles



Univer

sity o

f Oxfo

rd





Modelling for Cyclic Loading (0D): HARM (Abadie 2015)

0

0.2

0.4

0.6

0.8

0 0.1 0.2 0.3 0.4 0.5 0.6

No

rmalised

Ho

rizo

nta

l L

oad

Normalised Ground Displacement

0

0.2

0.4

0.6

0.8

0 0.1 0.2 0.3 0.4 0.5 0.6

No

rmalised

Ho

rizo

nta

l L

oad


H

e

D

L

Modelling Results from PISA (1D): HARM (Beuckelaers 2017)



Univer

sity o

f Oxfo

rd





Multidirectional Loading

Laboratory Model Tests (Richards 2019)



Univer

sity o

f Oxfo

rd





Realistic Storm Loading – Laboratory (Richards 2019)

Application of HARM to Storm Loading (Balaam 2020)

0

0.25

0.5

0.75

1

0 0.25 0.5 0.75 1

σcy

c/σ

REF

|σav|/σREF



Univer

sity o

f Oxfo

rd





• Develop new geotechnical design methods for offshore wind turbine foundations with a focus on – Cyclic loading including ratcheting effects

– Loading rate effects

– Robust calibration process

– Benchmark with field scale pile testing

• Translation into industrial practice for Ørsted

• Design optimisation for future wind farms

• Dissemination

Oxford – Ørsted Collaboration 2018-2023

WP1: Modelling

S

Mb

M

H

τ

τ

p

p

z

WP2: Calibration Methods

0

0.2

0.4

0.6

0.8

0 0.1 0.2 0.3 0.4 0.5 0.6

No

rmalised

Ho

rizo

nta

l L

oad


WP3: Theoretical Methods WP4: Field Testing

WP5: Application to Design




Univer

sity o

f Oxfo

rd






Non-Funded Research StudentsToby Balaam (WP2, WP3), Bohan Chen (CSC)

Iona Richards (WP2, WP3), Jonathan White (WP2)

Principal InvestigatorByron Byrne

Funded Research StudentsSarah Martin (WP4), Mark Qiu (WP2)Luc Simonin (WP3), Wayne Wu (WP2)

2 Students for October 2019

Post Doctoral Researchers / OthersStephen Suryasentana (WP1)PDRA to be appointed (WP4)

Clive Baker (Technician)

Oxford Academic TeamRóisín Buckley (WP2, WP4), Harvey Burd (WP1, WP3)Guy Houlsby (WP2, WP3), Chris Martin (WP2, WP3)

Ross McAdam (WP4), Brian Sheil (WP4)

Field Test Contractor(WP4)

To be Appointed

ØrstedJens Gengenbach (Contract), Amin Aghakouchak (Technical)

Chris Rogers (Contracts), Anne Petersen (R&D PM)

• Offshore wind energy is a major part of the UK Energy mix and will be for the foreseeable future

• Civil and geotechnical engineering is making significant contributions to offshore wind turbine design

• New approaches have been developed to more accurately design laterally loaded monopiles to allow site-wide optimisation

• Cyclic loading is the most significant challenge to address and is the focus of current efforts across the industry

• Oxford and Ørsted are collaborating to develop the next generation of design models to address cyclic loading

Concluding Remarks



Univer

sity o

f Oxfo

rd





The PISA Project was funded through the joint industry Offshore Wind Accelerator (OWA) program, designed and led by the Carbon Trust, with PISA Phase 1 supported by the UK Department for Energy and Climate Change (DECC) and PISA Phase 2 supported by the

Scottish Government.

Financial and technical support was provided by the following project partners: Ørsted (lead partner), E.ON, EDF, GE Renewable Energy, Iberdrola, innogy, SSE, Statkraft (Phase 1 only),

Equinor, Van Oord and Vattenfall.

The PISA Project Academic Work Group and numerous other people working for companies that supported the PISA Project.

Dr C.N. Abadie, T. Balaam, Dr W.P.A.J. Beuckelaers, I.A. Richards, Dr S.K. Suryasentana

For the Oxford-Ørsted collaboration the generous support of Ørsted, the UK’s Royal

Academy of Engineering and the University of Oxford is acknowledged.

Acknowledgements

Suction Installed Caissons – Guidelines – Out Shortly

𝐻

ℎ0𝑉0

2

+𝑀

𝑚0𝐷𝑉0

2

−2𝑎𝐻𝑀

ℎ0𝑚0𝐷𝑉02 − 4

𝑉

𝑉01 −

𝑉

𝑉0= 0

Mono-Caisson Multi-Caisson

L

Ds

L

D



Univer

sity o

f Oxfo

rd





Advanced Design Methods for Offshore Wind …...May 2016 Final report approved After May 2016...

Documents

Transcript of Advanced Design Methods for Offshore Wind …...May 2016 Final report approved After May 2016...