الفهرس 
chapter 1 : IntroductionOnly 14 pages are availabe for public view
 |  | from | 134 |  |  |
| | | from | 134 | | |
Abstract
In this thesis, optimal planning of primary distribution network using several distributed generations (DGs) types and sizes in optimal locations to minimize power losses is presented. The minimization of power losses leads to decreasing the cost of energy (COE), improving the voltage profile, enhancing power quality, supporting voltage stability, decreasing the effect of Carbone dioxide (CO2) produced from fossil fuels and increasing the lifetime of the network. Size and type of DGs are determined using HOMER software. Optimal location is achieved using Power World Simulator (PWS) and compared with that obtained from particle swarm optimization algorithm (PSO) and genetic algorithm (GA). Inserting distributed generations in distribution system affects the coordination of conventional protection devices.
The main protection available for distribution networks is non-directional over-current protective relays. Two of the main challenges that face that protection are inserting DGs and connecting the bus coupler. The first challenge (inserting distributed generations (DGs) to radial distribution feeders) may yield mis-coordination of the non-directional over-current protection. In order to overcome this mis-coordination, a proposed directional over-current protection approach for radial distribution feeders with/without DGs is introduced. All possible system topologies (with/without DGs) are simulated using genetic algorithm optimization to obtain relays’ settings achieving minimum operating time and to be suitable for all fault conditions. from this study, the worst case for coordinating each pair of relays is obtained. The second challenge of protective relays coordination in the distribution system is encountered due to availability of connecting the bus coupler during repairing fault periods. This will yield to additional errors in protection coordination due to changing fault current direction and magnitude. In order to overcome this challenge, a new approach to select proper relays’ settings that are suitable for all operating scenarios is applied using genetic algorithm considering all possible network configurations during normal operation and repairing fault periods.
The performance of the previous approaches is examined via simulating an 11-kV real primary distribution network located in Sadat city - Menoufia - Egypt (a new urban i
community) with latitude 30º 21.6´ N and longitude 30º 31.2´ E during normal operation and repairing fault periods. All applied test results have corroborated the accuracy of the proposed algorithm for clearing faults in distribution system.