Research Proposals and Collaboration Opportunities
If you are interested in developing any
of the following into a formal proposal for funding or in collaborating
with the proposer, please contact the proposer directly. ASRANet
will consider providing letters of support for proposals at the
funding application stage, if the proposal meets one of our priorities.
Click here to
see the priorities agreed by the Steering Group.
__________________
Research Project Topic, proposed by Dr Peter Gosling, Dept
of Civil Engineering, University of Newcastle
CREDO (Certainty,
Reliability & Engineering Design Optimisation) of Fabric
Structures
Summary
The development of a structural integrity methodology is proposed
for fabric structures. The aim of this proposal to ASRANet is
to facilitate collaboration between the fields of advanced structural
analysis and reliability analysis in an unusual area of structural
engineering. In this proposal the advanced structural analysis
research base is offered. Through ASRANet, the reliability expertise
is sought, thus fulfilling a primary objective of the network.
Essentially, CREDO aims to enhance the engineering solutions
to fabric structures through the novel, experimental, numerical
and analytical tools used to obtain those solutions, and to set
them in a reliability context. From this integrated approach,
professional and public perceptions of structural integrity,
equivalent to that attained by traditional, established techniques
will be obtained for fabric structures. With advances in material
specifications, engineering practice and information technology,
CREDO aims to provide a methodology by which structural engineering
and numerical research outcomes can be formalised. Both industry
and academic sources will be used. The analysis, design and construction
of fabric structures are used to illustrate the needs and difficulties
to be satisfied and surmounted as examples which may be extended
to solutions in other areas of non-conventional structural engineering.
Identifying the Needs and Current Methodologies of Fabric Structure
Engineering. In the 4th February 1999 issue of the New Civil
Engineer (NCE) a special feature appeared under the title "Creative
tension - wiring into tented structures". The main article,
written in the context of the Millennium Dome construction, opened
with the following - "Are giant tents domed? Tensile fabric
structures are among the most spectacular of the 20th century
- and the most controversial". This statement was made with
reference to the collapse of the new fabric roof over Montreal's
Olympic Stadium early last year (NCE, 28th January 1999).
Despite the increased and ambitious uses of architectural
fabrics as structural media, no clear consensus on the appropriate
factors of safety when specifying a membrane for a particular
project has emerged. For example, the current norm for architectural
fabrics is a factor of safety of 6. This is an unusually high
ratio stemming from earlier research on tear propagation principles.
A re-appraisal (potentially linked with work at Heriot Watt University,
Edinburgh) taking into account improved quality control in the
production consistency of fabrics, seam welding and physical
analysis may indicate a high level of redundancy.
The analytical tools currently used in the analysis of fabric
structures is based on a finite difference solution algorithm
to solve a continuum problem represented by a uni-dimensional
(i.e. constant elasticity) rather than a multi-dimensional (i.e
plane stress) discretisation. This methodology, developed in
the late 1970s-80s, is the sole approach used in the UK. It has
significant limitations in terms of mathematical rigour and robustness
and, as such, though effective for fundamental analyses of relatively
non-complex geometries, is limited in its application and extension
to necessary concepts such as reliability and optimisation. It
is proposed to reformulate the analytical model for fabric structures,
extracting positive features of the current approach, and setting
it within an appropriate mathematical framework.
No relevant design guidance currently exists in the UK for
the analysis and design of fabric structures. Consequently, "Industry
Standards" have arisen within the engineering and specialist
contracting community. These are typically based upon a working
load approach. Of all nations, including Japan and the USA as
pioneers of the application of glass-fibre architectural fabrics,
Japan is the only country to have a specific Building Law incorporating
requirements on factors of safety for structural membranes. However,
these requirements are about to change as all Japanese construction
codes are under revision to a limit state approach. In Germany,
a limit state philosophy is defined in the DIN standard for air-supported
structures (DIN 4134 - Traglufthallen) and is currently being
applied by some engineers in Germany to the broader design of
stressed membrane structures. The DIN 4134 standard addresses
some questions of seam and connection strength, ambient temperature,
environmental degradation and the nature of loading in an attempt
to provide a more rational basis for material selection and performance
prediction, but which remains dependent upon assumed independent,
superposed factors of safety.
Aim
· To provide a structural integrity (SI) methodology which
is appropriate to the optimal design of fabric structures.
Objectives/Deliverables.
· To develop a mathematically rigorous and robust (stable)
numerical algorithm for the analysis and design of fabric structures.
· Taking existing analytical, numerical and experimental
reliability theories and practices in the context of the integrity
of non-linear structures (typically geometrically non-linear),
to establish target SI optimality criteria, objective functions
and constraints.
· To propose a SI methodology consistent with the design
and physical behaviour of geometrically non-linear structures
founded upon reliability and optimisation theories and appropriate
structural mechanics.
· To disseminate that methodology through ASRANet, etc.
Relevance to other initiatives include: Technology Foresight
- within Construction, Reaping the benefits of the information
(technology) revolution and improved processes in construction
have been identified as major tasks; The European Commission,
Fifth Framework - Competitive and Sustainable Growth programme
states as main target to produce, disseminate and use the knowledge
and technologies needed to design and develop high quality products
which will be competitive on tomorrow's market; the DTI Teaching
Company Scheme. Cross-disciplinarity is reflected in the diversity
of the collaborators (material scientists, experts in structural
mechanics, structural engineers, fabric and engineering contractors)
and implicit through the multi-disciplinary natures of a number
of the participating organisations. For the chosen example of
fabric structure integrity the industrial collaborators are entirely
appropriate. This extends to the development and dissemination
of a generic SI methodology through the multi-disciplinarity
of several of the organisations. It is expected that any industrial
cash contributions will be used to raise a standard Ph.D. stipend
to case levels. CREDO will contribution to the success of the
managed programme by partially and fully achieving a number of
programme objectives within the information technology priority
area - a Ph.D. student and PDRA trained in SI; demonstrable in
industry performance; a change in SI methodology; introduction
of SI techniques to more industrial sectors and companies. Management
of the Project will be undertaken by a steering committee comprising
representatives from the research and industrial organisations
and the researchers, and chaired by an elected industrial partner.
Minutes of all meetings will be recorded. Actions will be by
consensus. Confidential disclosure agreements and collaborative
agreements will be implemented in the usual manner.
return to top of page
Proposal by Professor P.K. Das, Department of Naval Architecture
and Ocean Engineering, University of Glasgow and Professor M
Chryssanthopoulos, Department of Civil Engineering, University
of Surrey, Surrey for a Joint-Industry Project on
A Bench Mark
Study of Various Structural Reliability Analysis Software - A
Pilot Project
Aim
The objective of this project is to carry out bench mark work
of many structural reliability problems using various structural
reliability software. The problems will include various basic
structural components and applications to both onshore and marine
structures.
The software that will be considered in this pilot project
are PROBAN [4], STRUREL [3], CALREL [21] and SSMRA [5].
Other software like COMPASS [6] and NESSUS [7] may be considered
if more than 5 sponsors are obtained.
Scope of Work
1. Review and prepare a scheme for the development of a failure
functions directory and its potential role in improving safety
of structure either as a component or system basis.
2. Prepare a scheme for a basis for comparisons i.e. computational
efficiency, convergence criteria, handling of highly non-linear
failure surface, correlated variables, etc.
3. Selection of best state-of-the-art-formulation for the
following structural components:
· Beam
· Column
· Beam-Column
· Unstiffened Plate
· Stiffened Plate
· Stiffened Cylinders
In some cases, combined loadings will be considered.
4. Formulate problems for the following applications:
· Bridge structure
· Fatigue and fracture reliability both for onshore and
offshore structure
· Ship hull reliability
· Pipeline reliability involving operation and maintenance
· Submarine structure reliability
5. Establish statistical properties of basic design variables:
· Material properties
· Dimensions
· Initial imperfections
· Residual stresses
· Loadings (will be represented simple uncertainty)
· Resistance model uncertainty
6. Development of failure functions algorithm for the above formulations.
7. Carry out reliability analysis with the 4 softwares as
mentioned above using:
· Advanced First Order Second Moment (AFOSM) method
· Second Order Reliability Method (SORM)
· Directional Simulation (DIRS) with exact or approximate
surfaces
· Monte Carlo Simulation (MONT)
The failure surfaces will be based on the limit state function
of the relevant structural component. Correlations between design
variables will be taken into consideration in some cases.
8. Compile results according to the scheme established in
(1) and (2).
9. Prepare a final report and disk copy of the Directory.
return to top of page
PROJECTS proposed by Dr
Chun Q Li, Department of Civil Engineering, University of
Dundee
Topic One:
A Time-Dependent Reliability Tool for Benchmarking and Measuring
the Sustainability of Infrastructure
Examination of current research in structural assessment shows
that time-dependent reliability methods have not been widely
employed in determining risks of the structure. These methods
are necessary in risk assessment because almost all variables
involved in the problem behave as a stochastic process. This
project would be concerned with the development of a time-dependent
reliability tool that can be used to benchmark and measure the
sustainability of infrastructure and thereby to equip engineers
and asset managers of infrastructure with greater confidence
in making decisions regarding the maintenance and rehabilitation
of the infrastructure.
Topic Two:
An Information Adaptive Method for the Assessment of Existing
Structures Based on Bayesian Theory
Design loads for structures have increased significantly since
the advent of engineered structures with considerable excedance
of design and/or legal load on structures built in early years.
It is anticipated that both legal and design loads will continue
to increase in the future. The fact that many structures, although
old or maybe assessed to be "under-performed" based
on existing assessment methods, are still in service for many
years and perhaps performing reasonably well, has actually cast
great doubts on the accuracy and even the applicability of existing
methods for structural assessment. This project would be concerned
with the development of an information adaptive reliability method
for the assessment of existing structures with proven capacity
so that the load limit can be increased and thereby extending
the service life of the structure while the required safety level
is not compromised.
Topic Three:
A Framework for Whole Life Design of Concrete Structures
Current design codes, including the most updated version,
such as BS8500, only prescribe qualitative measures for durability
design. Although considerable research work has been devoted
to this area little progress has been made in quantitative durability
design to date. The reason appears to be that a framework for
developing rational design models for durability and a feasible
reliability methodology to implement the durability design is
still not in place. This project would be concerned with the
development of such a framework in which a suite of stochastic
models will be derived for the quantitative design of structural
durability. A reliability procedure would be formulated to determine
load and resistance factors widely used in design formulae so
that a linkage between the theory and practical application can
be established to ensure the facilitation with practitioners.
return to top of page
Project Proposals relating to Bridge Management, from Dr Victoria Hogg,
Highways Agency
Acceptance
Criteria
Acceptance criteria (i.e. target reliability index values) are
used and/or needed in most applications of reliability analysis
- including bridge design, bridge assessment and bridge management.
Various methods for the derivation of acceptance criteria exist.
Codes tend to use targets derived from values inherent in existing
codes. Other approaches include consideration of an acceptable
risk criterion and a socio-economic criterion. However, the acceptable
risk approach (and, depending upon the assumptions attached it,
the socio-economic approach) lead towards "absolute"
target values, whereas general reliability analyses result only
in "notional" values of the reliability index. The
question arises as to whether notional values can be compared
with these absolute acceptance criteria. The present lack of
guidance on this topic means that an element of engineering judgement
is required to relate reliability index values with acceptance
criteria.
If reliability analysis is to be accepted as a tool for use
in the assessment and/or management of bridges then robust acceptance
criteria need to be derived. A project is therefore proposed
to investigate and compare methods of deriving acceptance criteria
for use in bridge assessment and bridge management and address
the engineering judgement criteria surrounding their use.
FORM (iteration)
algorithm to ensure location of the global minimum design point
Recent applications of reliability analysis techniques in the
field of bridge management have highlighted difficulties in identifying
the global minimum reliability index value within FORM analyses.
A study aimed at determining latest intervention dates for essential
bridge maintenance utilised two different solution algorithms
(the Rackwitz-Fiessler iteration algorithm and NLPQL optimisation
algorithm). The results showed significant differences in both
the reliability index and sensitivity analysis results, depending
upon which algorithm was used.
The authors of a report on this study stress that if engineers
are to be expected to use reliability methods in their assessment
and/or management of real structures, then they need to feel
confident about results they obtain. This need for confidence
applies particularly to the global minimum result.
A project is therefore proposed to produce either advice on
the most appropriate (existing) algorithm, or a new, more robust
algorithm for use within FORM analyses.
Quantification
of sources of uncertainty in bridge management activities
There is a need to quantify sources of uncertainty in bridge
management activities and to target funds at addressing them.
For example, errors in inspection and testing techniques, model
uncertainty factors, human error etc. Additionally, there is
a need to identify levels of uncertainty associated with results
obtained from inspections and assessments. Uncertainty analysis
techniques, including fuzzy set analysis and Bayesian updating
are two techniques which can be used to quantify the level of
uncertainty attached to various bridge management activities.
As such, they can be used to provide some complementary measures
for consideration against any other deterministic or probabilistic
measures of condition or safety.
A project is proposed to identify the significant areas of
uncertainty within bridge management and also to develop techniques/methods
to address these uncertainties. The project could also investigate
more fully the potential for application of uncertainty analysis
techniques within bridge management. Within this, the project
could/should aim to develop complementary measures of uncertainty
to supplement existing measures of condition and/or safety. It
is hoped that the use of these measures would result in a more
clear understanding and interpretation of the results of reliability
analyses and therefore greater acceptance of its use from within
industry.
Reliability-based
design and assessment criteria for the SLS
Serviceability limit states can govern for the both design and
assessment of certain structure types and yet the derivation
of new reliability-based codes has (to date) concentrated on
the ultimate limit state.
A project is therefore proposed to provide a basis for carrying
out reliability-based design and assessment for the serviceability
limit state. The project would formulate reliability-based serviceability
limit state functions and design and assessment parameters. The
serviceability functions would rationalise acceptability limits,
which can be subjective.
(Note:- Project proposed by member of HA Steering Group for
Bridge Safety and Reliability. Internal HA funds already sought
for this project - no indication of funding as yet though.)
Extreme load
events and their combinations
Extreme load events such as scour, vehicle collision and high
winds require a special approach in design and assessment. There
is therefore a need to determine rational design and assessment
criteria for these events. In particular, there is a need to
determine the design forces for loads and their combinations,
which will provide an optimum reliability level for bridges.
A project is therefore proposed to provide a practical reliability-based
risk approach to design and assessment for extreme events.
(Note:- Project proposed by member of HA Steering Group for
Bridge Safety and Reliability. Internal HA funds already sought
for this project - no indication of funding as yet though.)
return to top of page
Projects suggested by : Dr Wangwen Zhao, Amey
VECTRA Limited
· Reliability of Oil/Gas Production
Systems
· Probabilistic Analysis of Explosion Response
· Probabilistic Risk Assessment for Seismic Structures
under Reduced Exposure
Proposal from Professor
Iain MacLeod, Dept of Civil Engineering, University of Strathclyde:
It is proposed that ASRANet promotes a holistic approach to
risk management in structural engineering. Issues beyond reliability
and advanced structural analysis include:
· Performance records.
It is proposed to establish a Scottish Centre for Forensic Engineering.
Support for this from all branches of engineering is sought.
· Criteria for robustness
· Cognition related risks
return to top of page
Proposals from Professor
Marios Chryssanthopoulos, University of Surrey
1) Probabilistic
Structural Integrity
Particularly probabilistic deterioration and assessment and the
use of structural monitoring for integrity assessment
(2) 'Intelligent'
Databases
How to set up databases that can be updated with new data in
an 'automated' way; for example, where new data is generated
by structural monitoring.
Research Needs
as Identified from questionnaires distributed at the ASRANet
Colloquium held December 2000, RINA, London. If you would like
to pursue these outline ideas further, contact ASRANet.
60 questionnaires were distributed, 14 returned, of which
6 said what research would help with their organisation's work.
Topics:
1. Ultimate strength and reliability of hull-girders (Bluewater)
2. Fatigue capacities and reliabilities (Bluewater)
3. FM assessments (Mott MacDonald)
4. Limit load solutions (Mott MacDonald)
5. Data Collection (Mott MacDonald)
6. Fast/easy access to data/databases and codes/standards (PAFA
Consulting engineers)
7. Review of current aim(?) reliability levels in different structures/industries
and how consequences are quantified. (Corus RD&T)
8. Practical examples of probabilistic fatigue analysis (worked
examples - SN & da/dN)
9. Response surface method (Univ Bristol)
10. Structural reliability methods (Univ Bristol)
11. Evaluation of uncertainty at early stage of design (Univ
Bristol)
return to top of page