Coastal Modeling - PowerPoint PPT Presentation

Loading...

PPT – Coastal Modeling PowerPoint presentation | free to download - id: 7220b4-ZWU3N



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Coastal Modeling

Description:

J.W. Kamphuis Coastal Modeling Tool. 1. Coastal Design. September 2010. 2. The Learning Curve. 3. How do we proceed? – PowerPoint PPT presentation

Number of Views:188
Avg rating:3.0/5.0
Slides: 65
Provided by: queensuCa
Learn more at: http://www.civil.queensu.ca
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Coastal Modeling


1
Coastal Modeling Indispensable Design tool
  • J. W. Kamphuis
  • Queens University
  • Kingston, ON, Canada
  • K7L 3N6

This paper will be posted on your website
2
1. Coastal Design
Coastal Modeling Indispensable Design tool
2. The Learning Curve
3. How do we proceed?
4. Integration
3
1. Coastal Design
  • Complex
  • Use of Models
  • Uncertainties

4
1a. Complex
Coastal Design
  • Processes are difficult to comprehend and not
    clearly understood.
  • Bottom shear stress
  • wave impact and energy dissipation
  • erosion, accretion
  • transport of sediment, pollutants, nutrients.
  • Designs are subject to difficult combinations of
    inputs
  • water levels, waves, tides currents and wind
  • Model outputs are difficult to interpret
  • morphology, environmental impact, water quality,
    etc.

5
Coastal Design
1b. Use of Models Traditional Design
Desk Study
Interpretation
Knowledge (Theory and Experience), Prototype Data
Preliminary Design
Modeling
Design
Post- Implementation Monitoring
Implementation
Trial and Error !
6
Coastal Design
Contemporary Design
Knowledge (Theory and Experience) and Prototype
Data
Post- Implementation Design
Preliminary Design
Preliminary Design
Preliminary Design
Modeling
Modeling
Modeling
Design
Implementation
Approval
Trial and Error !
7
Coastal Design
Models A tool for design optimization through
trial and error
8
1c. Uncertainties
Coastal Design
  • The reason for the trial and error is the high
    uncertainties in
  • Data
  • Understanding of coastal processes
  • Design by simply combining existing knowledge
    with data (desk study in the figure) can only
    produce very approximate answers conservative
    designs.

9
Coastal Design
Uncertainties
  • To improve design, uncertainties must be reduced
    in three areas
  • data,
  • understanding of the coastal processes
  • modeling techniques.
  • Moreover, clients, stakeholders, legislators,
    lawyers, etc. continue to press for (essentially
    zero) uncertainty in our design and in our
    understanding of the environment.

10
Coastal Design
Uncertainties
  • Further, acceptable levels of uncertainty have
    been reduced rapidly with time and hence
    reduction of uncertainties has become urgent.
  • But, is major reduction of uncertainty even
    possible? Also, it is a difficult task, since
    our tools were developed for an earlier time
    (BL)1 when uncertainties were accepted as part
    of life
  • Before Lawyers

11
Growth of Uncertainty
Coastal Design
12
Model Domains
Coastal Design
Traditional Models
13
Model Domains
Coastal Design
New Models
14
2. The Learning Curve
15
Learning (or Development) Curve
Oops !
Development (Knowledge, Religion, Business)
Rapid Progress
Time
16
Ages of the Learning Curve
Learning Curve
Old age
Development (Knowledge, Religion, Business)
Maturity
Infancy
Time
17
The Learning Curve of Knowledge
Learning Curve
Postmodern
Modern Era
Knowledge
Enlightenment
Yes. we can !
Time
1700
2000
1600
1800
1900
18
Ages of Knowledge Learning Curve Kuhn (1977)
Science becomes subculture (talks to itself) Work
is addressed to peers and adjudicated by
peers Challenges are internally
imposed (Improvement of theories, validating
paradigm)
Old age
Knowledge
Pressing problems are solved Development of
sophisticated theories Paradigm is articulated
Maturity
Solution of pressing practical problems Much
empiricism
Infancy
Time
19
Paradigm Shift
Learning Curve
Paradigm Shift (Sharp Break with the Old)
Development (Knowledge, Religion, Business)
Time
20
Learning Curve - Photography
Learning Curve
Digital
Colour
Photography
Film
Plates
Paradigm Shift
Time
21
Possible Decline
Learning Curve
Paradigm Shift
Development (Knowledge, Religion, Business)
Knowledge and development can decline after shift
Time
22
Coastal Learning Curves ??
Today
1970
Development ( of potential)
100
Physical Modeling
Numerical Modeling
75
50
Process Knowledge (Theory)
Field Data Collection
25
Time
0
1910
1950
1970
1930
1990
2010
23
Learning Curve
  • 1970
  • Learning curves are all quite steep, except for
    the learning curve for Physical Modelling. But
    it had produced the major tool of coastal
    engineering design.
  • The steepness of the numerical modeling, process
    knowledge and field measurement curves resulted
    from the introduction of computers.

24
Learning Curve
  • 1970
  • To advance in 1970, it made sense to concentrate
    on the steepest learning curves. Numerical
    modeling, process theory and field measurement
    methods and instrumentation were developed at the
    cost of physical modeling.
  • We took full advantage of the opportunities
    provided by the computer.

25
Learning Curve
  • 2010
  • All learning curves are quite flat.
  • This is the Old Age of Coastal Engineering
  • Therefore, we can not - must not - continue along
    the old paths followed since 1970.

26
Another View of Coastal Learning Curve ?
Paradigm Shift ?? (Introduction of computers)
Development (Knowledge, Religion, Business)
Decline in knowledge and development of Physical
Modeling
Time
27
3. How do we proceed?
28
Proceed?
  • Without question, the Numerical Model has become
    the tool of choice in todays Coastal Engineering
    design.

29
Proceed?
  • Rapid advancement can only occur through a
    paradigm shift
  • we should be so fortunate!

30
Proceed?
  • In the meantime, to advance at all (to be able to
    deal with the present and future design
    complexities, uncertainties and the approvals
    process) we need a concerted effort on all fronts
    process knowledge (theory), physical modeling,
    numerical modeling, field measurement.

31
Proceed?
  • We must take advantage of the particular
    strengths of each element.
  • We must integrate science and engineering,
    re-integrate theory and practice and integrate
    all our tools, people and facilities.
  • We must have (more?) open, frank communication
    between silos of expertise and generate (more?)
    mutual appreciation, understanding and help.

32
Proceed?
  • To be able to solve practical problems, we must
    eventually (further?) break down the various
    expert silos and revert to more generalist
    coastal engineering.
  • Very difficult, since career advancement is
    generally based on specialization, publication,
    etc.

33
4. Integration(of expertise, tools and people)
34
Integration
  • 4a. Closer (physical) integration of cultures
  • 1. Theory ? Practice

35
Integration
(physical) integration of cultures
  • 2. University education ? engineering
  • Define what we want in an engineering education
    theory application problem solving skills.
  • Interaction on the shop floor Students and
    Professors spend time regularly in industry
    Scholarships and sabbaticals in practice, etc..
  • Get engineers into the universities Engineers in
    Residence, Educational leaves, etc.
  • Industrial Academies e.g. Coastal software
    courses run by the software developers are a
    great idea to be further explored. Such courses
    must to be technically broad and teach theory as
    well as skills.

36
Integration
(physical) integration of cultures
  • 3. Physical ? Numerical Modelers
  • This is happening.
  • 4. Physical Modelers - Join forces, co-operate,
    share facilities and expertise.
  • HYDRALAB is an excellent example.
  • Problematic because of intellectual property and
    perceived leading positions.

37
Comment
Integration
(physical) integration of cultures
  • This integration will cost money.
  • Therefore, we must immediately be able to show
    added value
  • Better, more relevant education,
  • More competent engineers,
  • Shorter project approval periods,
  • Academic rewards for engineering as well as
    research,
  • etc, etc, etc.

38
Finally Integrated Modeling
Integration
  • Numerical models are it!
  • These models are usually are quite integrated
    with theoretical development.
  • But the users are not!
  • Numerical models need to be more integrated with
    data acquisition and with physical models
    (another type of data acquisition)
  • Hybrid Modeling.
  • What do we mean?

39
Usual Interpretation
Integration
Hybrid Modeling
Output
40
My Interpretation
Integration
Hybrid Modeling
Output
41
Integration !
Integration
Hybrid Modeling
42
Comment
Integration
Hybrid Modeling
  • Our models must be validated properly.
  • If this cannot be done by Field Measurements
    alone, calibrate and verify with results from
    large process experiments.
  • Improve Field Measurements (more and better data)
    by
  • Reduction of mobilization and other costs
  • Making experiments more transportable
  • Use new or different technologies
  • Develop some common standards

43
ThereforeHybrid Modeling Field Measurements
Physical Modeling Numerical Model
Computational Module
Integration
Hybrid Modeling
  • Level 2000 Models
  • Composite Models

44
Level 2000 Model
Integration
  • Far field is modelled numerically.
  • Near field is NM or PM with far field introduced
    as boundary conditions
  • Any PMs are of small prototype sections
  • Very Large Physical Models (VLPM) can achieve n
    1 to 5.
  • n 1 to 5 is already possible in oscillating
    tunnels and wave flumes we need a basin.

45
Integration
Level 2000 Model
Model Output
B.C for PM
B.C from NM
Input
Output
Far Field Numerical Model
Near Field Physical Model
B.C for VLPM
B.C for NM
Output
Near Field VLPM
Near Field Numerical Model
B.C for VLPM
46
Very Large Physical Models (Aside)
Integration
Level 2000 Model
  • Needed to
  • Advance understanding (scientific justification)
  • Reduce scale and laboratory effects
  • Provide Field Measurements under controlled,
    repeatable conditions
  • But
  • Who will build them?
  • Are they economically justified?
  • Commercially not viable !

47
And.
Integration
Level 2000 Model
  • VLPM experiments are like field experiments
  • Extensive planning and long lead times
  • Collective exhaustion and recovery of year(s)?
  • Therefore extensive downtime of facility
  • Financing for
  • Experiments?
  • Facilities?
  • Downtime?

48
Therefore.
Integration
Level 2000 Model
  • We want only one or two excellent, world class
    VLPM facilities (rather than a number of
    inadequate or barely adequate, locally financed
    facilities).
  • Given the economic environment and recent
    history, the success of this approach is in
    doubt.
  • What else can we do?

49
Composite Model
Integration
Output
Model Results
B.C.
50
Integration
Composite Model
  • Process models are
  • Physical or numerical
  • models must be trustworthy representations
  • simple, inexpensive, easy to understand tests
  • generic tests and test results
  • repeatable results
  • Process model results form the bricks

51
Integration
Composite Model
  • Computational module forms the mortar
  • not necessarily a numerical model
  • Mostly, simple accounting of results
  • Calibration is done here (inexpensive)
  • What if? Is done here (inexpensive)
  • Not necessarily same provider for two parts
  • We could develop a world-wide Bank of process
    model bricks (I wish)

52
Integration
Composite Model
  • Example Sacrificial Beach Drilling Islands in
    the Arctic Ocean, 1980
  • Ice is Nice
  • 9 mo. instead of 2 mo.
  • No Protection

53
The Process Model
Composite Model
Integration
Arctic Island Models
54
The Process Model
Composite Model
Integration
Table of Test Parameters
55
Composite Model
Integration
56
Composite Model
Integration
57
Composite Model
Integration
58
Composite Model
Integration
59
Composite Model
Integration
60
Composite Model
Integration
61
Composite Model
Integration
62
The Process Model
Composite Model
Integration
Island Models
63
The Computational Module
Composite Model
Integration
Island Model Calibration
Field !
64
This paper will be posted onwww.civil.queensu.ca
Thank You
About PowerShow.com