Akademik Veri Yönetim Sistemi
Tezin Türü: Doktora
Tezin Yürütüldüğü Kurum: Marmara Üniversitesi, Mühendislik Fakültesi, Endüstri Mühendisliği Bölümü, Türkiye
Tezin Dili: İngilizce
Öğrenci: İlayda Ülkü
Danışman: Çiğdem Alabaş Uslu
Özet:
Wind
energy is firmly favoured worldwide as a renewable energy resource for
electricity generation. Due to the climate impacts of offshore wind farms, the
number and scale of offshore wind farms have been increasing. Design of a wind
farm invokes several optimization problems from the selection of wind farm
location where wind regime is efficient. A wind farm, mainly, is composed of a
set of turbines, one or transmitters and a set of electrical cable connections
between turbines and transmitters. The problem considered in this study has
three main parts: location of the turbines and connection of the located
turbines that is cable layout design and a simultaneous solution for both turbine
location and cable layout problems.
Determination
of locations for each turbine within the wind farm is called turbine location
(TL) problem. In the first part of the thesis, a novel nonlinear mathematical programming
model for the layout of wind turbines considering the multiple wake effects is
proposed. In the model two objective functions are taken into account
separately: maximization of total power produced and minimization of cost for per
unit of the power. Relative positions of turbines affect the amount of total
energy generation because of wake effects. Jensen's wake decay model is used to
involve multiple wake effects into the proposed model. It was proven that solution
space of the proposed model has the totally unimodularity property and
relaxation of binary variables which show the wind turbine locations makes the
model relatively easy to solve. Experimental study for the TL problem of the
thesis exhibits that total power production and cost per unit of power generated
from the developed model outperform that of the previous existing studies in
the literature on a suit set of case problems and thus, can be employed to
layout more productive and cost effective wind farms.
The
determination of cable connections among turbines and transmitters to collect
energies produced by turbines at one or more transmitters is called cable
layout (CL) problem. In the second part of the thesis, cable connections
between turbines and transmitter(s) are considered that is one of the
challenging problem in the design of wind farm layout. This problem, called CL
design, is especially important for offshore wind farms where the featured and
highly costed cables are used. The second part of the present study focuses on
the optimization of cable connections of offshore wind farms to minimize the
total cost of cables. The problem is modelled as a mixed integer linear program
(MIP) respecting different cable types and capacities, non-crossing
connections, and a limited number of total connections to a single transmitter
depending on the transmitter type. Since the design of CL is an NP-hard
problem, MIP model is modified by introducing additional heuristic rules, which
are specific to the problem, to enhance the capability of MIP in solving of
large offshore wind farms. An experimental study conducted on a suit set of
test problems and real-life cases reveals that the proposed optimization models
are able to find optimal or good feasible solutions within a reasonable amount
of computer time compared to the previous studies in the literature.
In
the third part of the present study, TL and CL are considered simultaneously. A
MIP_COMBINED approach
for offshore wind farms to solve both wind TL and CL problems. A new linear
integer mathematical model is developed to deal with simultaneously solving of
wind TL and CL problems. A set of test instances are used to show the
performance of the proposed model. Experiments showed that both linearization
of the objective function and binary variables of TL and CL problems makes
MIP_COMBINED model hard to solve. Only three test cases out of twelve cases are
solved by the MIP_COMBINED model. However, sequentially solving of TL and CL
problems considering the power function in the TL part provided better cost
effective solutions comparing the MIP_COMBINED model.
Since
the wind energy field of research has relatively great interest, solving the
design problems takes an important part to generate new ideas. In this
challenging context the present thesis focuses on the optimal design of wind
farm layout and of cable routing simultaneously.