الفهرس 
صفحة العنوانOnly 14 pages are availabe for public view
 |  | from | 183 |  |  |
| | | from | 183 | | |
Abstract
Nowadays, for technical and economic considerations besides
environmental awareness, the application of distributed generators (DGs)
such as photovoltaic, wind, hydrogen fuel cells, microturbine, diesel, etc. is
increasing. The local interconnection between these distributed generators
and load is called microgrid (MG). This microgrid can offer many benefits
such as losses reduction during energy transfer from generation unit to
load, reduction the amount of harmful gases (greenhouse gases GHG ) by
utilization the green energy resources such as wind and photovoltaic and
enhancement the system reliability due to dependence on multi-energy
resources. Moreover, the islanding operation of these MGs improves the
system physical and cyber security.
However, in order to enable the microgrid to perform its function,
there are some challenges should be faced and overcame. These challenges
include maximization the utilization of renewable distributed generator by
tracking the maximum power point. In addition, dispatching the power
among the different DGs in order to reduce the generation cost and
greenhouse gases (GHG) emission. Also, design suitable control
mechanism to maintain the system stability in normal and abnormal
conditions such as intermittent in the available power from renewable
resources or outage one of generation units. Moreover, the microgrid ability
to expand by adding new generation units.
The Thesis suggests a novel technique to maximize the capture power
from Photovoltaic (PV) system by using an artificial neural network
(ANN). The proposed technique achieves an excellent tracking factor,
eliminates the needing of irradiance devices. Consequently, it is costeffective. Moreover, it does not suffer from the mismatch between iii
irradiance measuring devices and actual irradiance strikes the PV array.
Eventually, the proposed technique is capable of tracking the MPP
irrespective of load nature.
The thesis investigates the development of an optimal droop
controller to minimize the total generation cost. The suggested controller
guarantees the economic operation in case of generation outage, load
variation, and PV output variation. In addition to it does not require
forecasting of the load and the PV profiles. Moreover, the proposed droop
controller considers the effect of the distribution system on power haring
accuracy through maximizing the droop gains. Also, the thesis presents an
optimal droop controller to minimize the amount of GHG emission.
Eventually, in order to enhance the scalability of the microgrid, the
thesis suggests robust static state feedback (SSF) local controller based on
H2/H∞/pole-placement/polytope. This controller is robust against generation
outage, constant power loads, structured uncertainty and unstructured
uncertainties. This proposed tuning technique is compared with H2, mixed
H2/H∞, H2/H∞/pole-placement techniques and proved its superiority. The
proposed control