You are on page 1of 9

flexibility and its

effect^dn^"acket structures-
N^ / / / /
a pilot study on two-dimensional frames
WHAT IS UEG?
UEG is the research and information group for the
underwater and offshore engineering industries. Its
main function is t o provide a means of cooperation
between its members, including government, in solving
common technical problems, obtaining and sharing
information and in providing an industry-based focus for
research.
UEG is non-profit-making and its financial base is
provided by the annual subscriptions of its members.
Additional finance for individual projects is obtained
from industrial and government organisations in-
terested in any of the specific project areas.
Membership of the Group is open t o any organisation
wi t h an interest or involvement in underwater or
offshore engineering. Each member's subscription is set
in relation t o their size and involvement in underwater
or offshore engineering.
To ensure its industrial relevance, UEG's programme is
defined and selected by a number of committees
through whi ch member representatives are able t o put
forward their future research needs. Currently four
Working Groups advise the UEG staff on research
requirements in the fol l owi ng areas: Diving and Man
under Water: Offshore Structures: Underwater Engineer-
i ng: and Maintenance of Offshore Installations.
The UEG Committee determines the Group's policy and
overall programme and authorises expenditure on
individual projects. The Committee is responsible t o the
Council of UEG's parent organisation CIRIA (the Con-
struction Industry Research and Information Associ-
ation) and operates wi t hi n limits set by the Council.
UEG projects are managed by the full-time staff and
industrial involvement is provided by Steering Groups
wi t h membership drawn from the relevant sector of the
offshore industry.
UEG implements its projects by placing contracts wi t h
those best able t o undertake the work, and has no
laboratory facilities of its own. By avoiding the con-
straints and long-term commi tment of specialised staff
and laboratories, the Group offers the flexibility of
operation required t o undertake research work associ-
ated wi t h the changing needs of t he offshore industry.
The results of all UEG projects are published in reports
which are issued free of charge t o members. Selected
reports are later sold at a very much higher price t o
non-members. The proceeds of additional sales are used
t o assist in financing the future research programme.
In addition t o contact wi t h its members. UEG maintains
links wi t h many other research and technical organi-
sations concerned wi t h offshore and underwater engin-
eering. It is the official channel for the release t o industry
of the Royal Navy Diving Tables and other related
information.
The activities of UEG are outlined in the UEG Annual
Report available free of charge from the address below
Requests for further information about UEG, including
enquiries about membership, should be sent t o the
Manager, UEG, 6 Storey's Gate, Westminster, London
SW1P3AU
REPORT UR22 1984
Node fl exi bi l i ty and its effect on j acket structures
a pi l ot study on t wo-di mensi onal frames
Price 72 (18 UEG Members)
I CI RI A 1984
ISSIM: 0305 4055 I SBN: 0 86017 213 9
6 Storey's Gate Westminster London SW1P 3AU Telephone 01-222 8891
THE RESEARCH AND INFORMATION GROUP FOR THE UNDERWATER AND OFFSHORE ENGINEERING INDUSTRIES
A PART OF CIRIA-THE CONSTRUCTION INDUSTRY RESEARCH AND INFORMATION ASSOCIATION
The project leading to this report was carried out under contract to UEG by W S Atkins &
Partners where the staff engaged on the work were N P D Barltrop BSc CEng MICE MRINA,
A Culverhouse BSc and MJ McLoughlin BEng CEng MICE MIStructE. The work was financed
through UEG by a joint venture of thirteen organisations including the Department of Energy. The
report has been prepared with the assistance of a Project Steering Group comprising representa-
tives ofthe sponsoring organisations and Professor Fessler ofthe University of Nottingham.
Mr R K Venables (Chairman) UEG
W S Atkins & Partners Mr N P D Barltrop
Dr C J Billington
Mr J Chattaway
Professor H Fessler
Dr S L Fu
Mr M B Gibstein
Mr B Marlow
Mr D McManus
Mr M.J McLoughlin
Mr P E G O'Connor
Mr M C O'Flynn
Mr A G Reynolds
Mr B Ringstrom
Mr R J Simpson
Mr S G Stiansen
Mr F E S West
Wimpey Offshore Ltd
Posford, Pavry & Partners
University of Nottingham
Gulf Oil Exploration and Production Company
Det norske Veritas
Marine Technology Support Unit
Shell UK Exploration and Production
W S Atkins & Partners
Amoco (UK) Exploration Company
Taylor Woodrow Construction Ltd.
BP International Ltd
Norwegian Petroleum Directorate
UEG, (Research Manager for the project)
American Bureau of Shipping
Harris & Partners
Acknowledgements
This work made extensive use of results, at the time unpublished, obtained from model tests
by Professor H Fessler and the late Mr H Spooner ofthe University of Nottingham.
These results enabled the authors to do much more work on the effects of flexible nodes on
structural behaviour than would otherwise have been possible.
The United Kingdom Offshore Steels Research Project have also allowed the use of their finite
element results so that flexibility matrices derived from finite element and model tests could be
compared.
UEG Report 22
Contents
Page
No.
LI ST OF TABLES 5
LIST OF FI GURES 5
SUMMARY 6
NOTATI ON 6
ABSTRACT OF REPORT, RESULTS AND CONCLUSI ONS 7
1.0 I NTRODUCTI ON
1.1 General 10
1.2 Previous Research and Sources of Data 10
2.0 DESCRIPTION OF WORK
2.1 The Frames 12
2.2 Joints 14
2.2.1 General 14
2.2.2 Joi nt Nomenclature 14
2.3 Loading 14
2.4 Buckl i ng Analysis 16
2.5 Natural Frequency Calculations 16
3.0 LOCAL FLEXI BI LI TY OF JOI NTS
3.1 General 17
3.2 Comparison Requirements 17
3.3 Eccentri ci ty 17
4.0 RESULTS
4.1 General 19
4.2 Frame Global Deflections 19
4.3 Relative Deflections wi t hi n Joi nts 20
4.4 Effect of Fl exi bi l i t y on Axi al Forces 21
4.5 Effect of Fl exi bi l i t y on Bending Moments 23
4.5.1 Choice of Out put t o Present 23
4.5.2 Summary of Bending Moment Changes in
Non-Eccentric Joints 30
4.5.3 Bending Moment Changes in Eccentric Joi nts Caused
by Fl exi bi l i t y Changes 31
4.5.4 Bending Moment Changes Between Eccentric &
Non-Eccentric Joints bot h Conventi onal l y Analysed 31
4.6 Effect of Joi nt Fl exi bi l i t y on Brace Buckl i ng 38
4.7 Effect of Joi nt Fl exi bi l i t y on Vi brat i on Characteristics 41
5.0 I NTERPRETATI ON OF RESULTS
5.1 General 63
5.2 Interpretati on of Defl ecti on Change 69
5.3 Interpretati on of Axi al Force Change 65
5.4 Interpretati on of Bending Moment and Bending Stress Change 65
':, 5.4.15.4.24 Discussion of the Vari ous Joi nts i n the
di fferent structures 65
5.5 Reasons for Differences in Bending Moment Changes Between
the Structures 69
5.6 Interpretati on of Buckl i ng Results 70
5.7 Interpretati on of the Natural Frequency Results 70
5. 7. 1- 5. 7. 3 Structures 1- 3 70
5.7.4 Effect of Natural Frequency Changes 70
UEG Report 22
6.0 ESTI MATI ON OF EFFECTS
6.1 General 71
6.2 Esti mati on of Axi al Force Change 71
6.3 Esti mati on of Defl ecti on Change 72
6.4 Esti mati on of Bending Moment Change 73
6.4.1 Bending Moment f r om Axi al Forces (Non-Eccentri c Joi nt) 73
6.4.2 Bending Moments f r om Transverse Brace Loading 74
6.5 Brace Buckl i ng Load Cal cul ati on 74
6.6 Natural Frequency Changes 74
7.0 RECOMMENDATI ONS
7.1 General 75
7.2 Structures Similar t o those Analysed in this Report 75
7.3 Structures Not Similar t o those Analysed in this Report 76
7.4 Out-of-Plane Fl exi bi l i t y Effects 77
7.5 Eccentri ci ty 77
7.6 Research requirements 77
APPENDI X A
Fl exi bi l i t y data used i n study 78
APPENDI X B
Details of t he deri vati on of the j oi nt stiffness matrices f r om the f l exi bi l i t y
data
B.l General 83
B.2 Possible methods of properl y al l owi ng f or local f l exi bi l i t y i n t he 84
j oi nt .
B.3 Details of deri vati on of j oi nt stiffness matri x f r om f l exi bi l i t y data 86
B.4 Deri vati on of t he addi ti onal f l exi bi l i t y terms t hat are required 89
B.5 Inversion of j oi nt f l exi bi l i t y t o j oi nt stiffness 94
APPENDI X C
Some comments on the various matrices
C.I Fl exi bi l i t y and stiffness matrices 97
C.2 Relationship between f l exi bi l i t y and stiffness matrices 104
C.3 Exami nati on of the matrices 104
C.4 Effect of eccentri ci ty i n the j oi nt 106
C.5 Comparison of MP1 matrices and conventi onal terms 106
APPENDI X D
Deri vati on of t he buckl i ng curves 107
APPENDI X E
Computer programs 109
APPENDI X F
Comparison of an experi mental l y produced f l exi bi l i t y matri x wi t h one 110
calculated f r om f i ni t e element results
APPENDI X G
A note on the comparison of structural responses t o dynami c loading 115
REFERENCES 116
UEG Report 22
List of Tables
List of Figures
Table 2.1
Table 2.2
Table 4.2.1
Tables 4. 4. 1. 1- 3
Tables 4. 4. 2. 1- 3
Tables 4. 4. 3. 1- 3
Tables 4. 5. 1. 1. - 3
Tables 4. 5. 2. 1- 3
Tables 4.5.3
Table 4.5.4
Table 4.6.1
Tables 4. 7. 1- 3
Table 7.1
Table 7.2
Table A. I
Table A. 2
Table C. I . 1
Table C.I .2
Table C.I .3
Joi nt types 15
Details of appl i ed l oad cases 15
Displacemen t results 19
Axi al forces St ruct ure 1 21
Axi al forces St ruct ure 2 22
Axi al forces St ruct ure 3 22
Bendi ng moment ampl i tudes MP1- CP1 - St ruct ure 1 24
Bendi ng moment ampl i tudes MP2CP1 - St ruct ure 1 27
Summary of bendi ng stress changes 30
Index f or bendi ng moment diagrams 31
Buckl i ng l oad changes 39
Comparison of natural frequencies (and i ndex t o mode shape 41
diagrams)
Summary of changes 76
Summary of changes caused by eccentri ci ty 76
Model numbers of experi mental investigation provi di ng f l exi bi l i t y 79
data
Fl exi bi l i t y data 79
Fl exi bi l i t y mat ri x f r om Professor Fessler 98
Expanded f l exi bi l i t y mat ri x i n Cartesian system 98
Mat ri x i nverted t o give stiffness mat ri x and chor d sub-matri x 98
values added
Figure 2.1
Figure 2.2
Figure 2.3
Figure 2.4
Figure 2.5
Figure 2.6
Figure 4.2.1
Figure 4.2.2
Figure 4.2.3
Figure 4.3.1
Figure 4.3.2
Figures 4. 5. 1. 1- 8
Figures 4. 5. 2. 1- 8
Figures 4. 5. 3. 1- 8
Figures 4.6.1
Figure 4.6.2
Figure 4.6.3
Figures 4.7.1.1 - 13
Figures 4. 7. 2. 1- 13
Figures 4. 7. 3. 1- 13
Figure 8.1
Figure B.2
Figure 8.3
Figure 8.4
Figure 8.5
Figure 8.6
Figures C. I . 1- 10
Figures C.I . 11- 20
Figure D.I
Structure 1 12
Structure 2 12
Structure 3 12
Fi ni t e element model of Structure 1 13
Fi ni t e element model of St ruct ure 2 13
Fi ni t e element model of Structure 3 13
Defl ected shape of St ruct ure 1 20
Defl ected shape of Structure 2 20
Defl ected shape of St ruct ure 3 20
Structure 1 j oi nt defl ecti ons 20
Structure 2 j oi nt deflections 20
Bending moment diagram Structure 1 32
Bendi ng moment diagram Structure 2 34
Bendi ng moment diagram St ruct ure 3 36
Buckl i ng l oad changes St ruct ure 1 38
Buckl i ng l oad changes Structure 2 39
Buckl i ng l oad changes St ruct ure 2 40
Mode shape comparisons Structure 1 43
Mode shape comparisons St ruct ure 2 49
Mode shape comparisons Structure 3 56
Al t ernat i ve j oi nt modes f or two-brace j oi nt 84
Type of stiffness mat ri x resul ti ng f rom each model 85
77?e degrees of freedom of the experi mental work 86
The degrees of freedom requi red i n the j oi nt stiffness mat ri x 87
Loads and relative displacements on a two-braced j oi nt 87
Loads and absolute deflections on a two-braced j oi nt 94
Di agrammati c representation o f stiffness and f l exi bi l i t y terms
f or the Fessler 15 - MP1/ ME1 j oi nt 99
Diagrammatic representation of stiffness and f l exi bi l i t y i mpl i ed
by conventi onal analysis 99
Relationship between and rot at i onal stiffness and buckl i ng l oad
f or an axi al l y compressed beam 109
UEG Report 22
Summary
This Report presents an investigation ofthe commonly ignored effect of chord wall flexibility
at brace connections on the behaviour of oil production jacket structures.
It considers the effects on the in-plane deflections, axial forces, bending moments, brace
buckling and natural frequencies of three 100 m tall vertical plane frames. The effects of out-
of-plane loading and joints with braces in more than one plane were outside scope of the study.
Results from conventional analysis, in which the braces are extended to connect rigidly with
the chords at their centre line intersection points, are compared with other analyses in which
the joints are represented by a stiffness matrix. Approximate 'hand' methods for assessing
the effect of joint flexibility on other structures are suggested.
Notation
A
A
B
C
d
D
e
ep
E
f
F,f
g
G
H
H
I
k
K,k
L
Lb
Le
M
M
P
Pcrit
t
t
T
T
u, v
x,y,z
x
x
X
X, Y, RZ
a
P
y
5
5
V
e
V
a
0
^
, ^
measure of chord wall bending stiffness/brace bending stiffness (End 1)
cross-sectional area
measure of chord wall bending stiffness/brace bending stiffness (End 2)
compressive forces
diameter of braces
diameter of chords
eccentricity of brace intersections on chord centre line
eccentricity of brace intersections perpendicular to chord centre line
Young's Modulus
flexibility
flexibility matrix
brace-brace gap
modulus of rigidity
height of wave
equilibrium matrix
inertia
stiffness
stiffness matrix
length
length (between chords)
length (effective)
moment
mass matrix
force
7r2EI/L2
brace thickness
time
chord thickness
tensile force
lengths
lengths of co-ordinates
response
vector of displacements of nodes on a structure
vector of accelerations of nodes on a structure
nodal freedoms in plane frame analysis
chord wall rotational stiffness
chord wall rotational stiffness
brace-chord angle
deflection
change
reduction factor on bending stress
angle or rotation
Poisson's Ratio
stress
angle or rotation
brace direction freedom
frequency
UEG Report 22
Abstract of report, results and conclusions
INTRODUCTION
Jacket type offshore structures are conventionally analysed as frames with members connected
at rigid joints.
In fact the joints are not rigid.
This report is concerned with the errors in deflections, nominal stresses, buckling load and
natural frequencies involved in assuming rigid joints. The nominal stress excludes any stress
concentration factor.
The study is limited to in-plane effects in vertical frames.
SCOPE OF WORK
A method was developed for usingjoint flexibility data, including the important brace to brace
coupling terms, within a jacket analysis.
A number of jacket frames were then analysed both conventionally and with flexible joints.
The results were compared numerically and qualitatively. Recommendations were made con-
cerning analysis and design.
THE ANALYSIS
Stiffness matrices representing brace-chord and brace-brace in-plane connectivity, for various
tubular joints, were calculated from flexibility data made available by Professor Fessler O of
Nottingham University. The data in the form of flexibility matrices covered the required range
of T-joints but only the standard D/T = 25.3, d/D = 0.53 K-joints. To estimate the required
matrices for the other K-joints it was necessary to obtain flexibility ratios from the T-joints and
to apply those to the available K-joint values. A YT-joint with D/T = 50.6 and d/D = 0.53 was
also available.
Three structures representative of North Sea jacket frames were each analysed with seven
different types of joint.
Structure 1 Structure 2 Structure 3
Main chords were 1700 x 30; inclined braces 900 x 25;
horizontal braces 750 x 25.
UEG Report 22