الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 208 |  |  |
| | | from | 208 | | |
Abstract
Partitioning of large scale distribution systems into multiple Microgrids (MG) is considered as an effective solution to maintain systems reliability and to improve its restoration under emergency conditions. The partitioning process faces many challenges especially if the uncertainty nature of both renewable energy resources (RERs) and loads, and the unbalanced operation of the distribution systems are considered.
Power flow calculations are an essential tool for efficient designing, planning or operating process of radial distribution systems. In this thesis,
a Probabilistic power flow method is proposed for analyzing unbalanced radial distribution systems. The proposed method for power flow is a suitable, fast and accurate method for analyzing unbalanced, complex, passive and active distribution systems. Also, it is characterized as an effective convergence method for complex distribution system that includes distribution system devices such as distribution transformers and voltage regulators, . . .etc. Moreover, the proposed power flow improves the speed of analysis over the conventional methods by using the proposed coefficient matrices. In addition, it considers the uncertainty nature of the integrated renewable energy resources and the load demand using normal probability density functions.
The proposed power flow method is conducted on IEEE 13-node, IEEE 34-
node, 37-node, and 123-node unbalanced distribution systems. RERs integration in power systems may results in operational and planning issues. These issues revolve around the RER uncertainty and its applications on probabilistic power system operation and planning. So that, a novel and effective criterion for modeling RERs and system loads uncertainties is proposed. Four sorting stages are applied to reduce the number of uncertainty cases. Three different uncertainty cases reduction strategies are proposed for obtaining different accuracy and speed options. The proposed reduction strategies are tested on medium and large scale distribution systems. Four phases for uncertainty reduction are applied based on the main sources of uncertainty such as solar irradiation, temperature, and wind speed. Moreover, the load uncertainty and system unbalance are considered. Also, this thesis proposes an effective method for the optimal planning of RERs in unbalanced distribution systems. Different types of RERs, as PV and Wind, are considered. A multi-objective optimization problem
is formulated using non-dominated sort and crowing distance for multi objective hybrid algorithm that combines the merits of Particle Swarm Optimization (PSO) and Sine Cosine Optimization (SCO) algorithms. Three objectives, improving voltage profile, power losses minimization, and minimization of voltage unbalance, are considered. The proposed method for RERs planning determines the optimal RERs’ specifications such as site/- size/type of DGs, number of PV modules or wind turbines in each DG and their power factor and connection phases. Also, it determines the optimal setting of substation tap changer and system voltage regulators. Moreover, both RERs and loads uncertainty are considered and modeled using Monte Carlo simulations. Finally, the active distribution system partitioning is handled through
a proposed three stages method. The proposed method for partitioning
presents a parallel planning for both energy sources integration process and
partitioning processes. It considered all the system circumstances. Also, it
determines the optimal specification of the devices that will be added to the
system to reach the target efficiency of the system during system partitioning and unification processes. The proposed three-stage partitioning method considers the uncertainty of PV modules and wind turbines. The proposed method for partitioning is tested on a large-scale unbalanced test feeder.