TY - BOOK

T1 - A deterministic combination of numerical and physical models for coastal waves

AU - Zhang, Haiwen

PY - 2006/8

Y1 - 2006/8

N2 - Numerical and physical modelling are the two main tools
available for predicting the influence of water waves on coastlines
and structures placed in the near-shore environment. Numerical models
can cover large areas at the correct scale, but are limited in
their ability to capture strong nonlinearities, wave breaking, splash,
mixing, and other such complicated physics. Physical models naturally
include the real physics (at the model scale), but are limited by the
physical size of the facility and must contend with the fact that
different physical effects scale differently. An integrated use of
numerical and physical modelling hence provides an attractive
alternative to the use of either tool on it's own.
The goal of this project has been to develop a deterministically
combined numerical/physical model where the physical wave tank is
enclosed in a much larger computational domain, and the two models
communicate via a multi-flap wave generator placed along one boundary
of the physical model. Previous work in this regard has typically
been by means of a one-way transfer of stochastic information based on
linear theory, which precludes the transfer of important phase and
nonlinear information. A new ad hoc unified wave generation method
has been developed which combines the theories of nonlinear shallow
water generation and linear deep water generation. The numerical
model used is the Mike21BW model developed at DHI - Water \&
Environment. The theory is tested in a wave flume (2-D waves) and in
a 3-D wave basin and is generally successful. The method is also
tested with the numerical model replaced by a fully nonlinear periodic
wave theory (Stream Function Theory) and it turns out to be far
superior to existing wave generation techniques in it's ability to
generate highly nonlinear periodic waves of constant form.

AB - Numerical and physical modelling are the two main tools
available for predicting the influence of water waves on coastlines
and structures placed in the near-shore environment. Numerical models
can cover large areas at the correct scale, but are limited in
their ability to capture strong nonlinearities, wave breaking, splash,
mixing, and other such complicated physics. Physical models naturally
include the real physics (at the model scale), but are limited by the
physical size of the facility and must contend with the fact that
different physical effects scale differently. An integrated use of
numerical and physical modelling hence provides an attractive
alternative to the use of either tool on it's own.
The goal of this project has been to develop a deterministically
combined numerical/physical model where the physical wave tank is
enclosed in a much larger computational domain, and the two models
communicate via a multi-flap wave generator placed along one boundary
of the physical model. Previous work in this regard has typically
been by means of a one-way transfer of stochastic information based on
linear theory, which precludes the transfer of important phase and
nonlinear information. A new ad hoc unified wave generation method
has been developed which combines the theories of nonlinear shallow
water generation and linear deep water generation. The numerical
model used is the Mike21BW model developed at DHI - Water \&
Environment. The theory is tested in a wave flume (2-D waves) and in
a 3-D wave basin and is generally successful. The method is also
tested with the numerical model replaced by a fully nonlinear periodic
wave theory (Stream Function Theory) and it turns out to be far
superior to existing wave generation techniques in it's ability to
generate highly nonlinear periodic waves of constant form.

KW - Deterministic combination

KW - Numerical model

KW - Physical model

KW - Coastal waves

M3 - Ph.D. thesis

SN - 87-89502-66-3

BT - A deterministic combination of numerical and physical models for coastal waves

PB - Technical University of Denmark

CY - Kgs. Lyngby

ER -