الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 174 |  |  |
| | | from | 174 | | |
Abstract
Recently, the numerical relays with the ability of communication are used to provide adaptive solutions for protection problems associated with the connection of Distributed Generations (DGs) into power systems such as loss of coordination, bi-directionality, and islanding detection. In this context, this thesis aims to present an adaptive protection scheme based centralized and decentralized communication combination to solve the protection problems imposed by the insertion of DG in power systems. Also, it is aimed to improve the detection of faults that are difficult to be detected by conventional relays whose value is not dependent on the connection of DGs such as High Impedance Faults (HIF). The proposed method uses two directional relays in each feeder; one of them at the beginning of the feeder (called forward relay) and the other at the end of the feeder (called reverse relay). Those relays are numerical relays with the ability of communication between each other and between them and the central unit. The communication process is based on IEC 61850 communication standard which enables the integration of all protection, control, measurement and monitoring functions. ETAP program will be used to simulate the tested system and to perform offline load flow analysis and short circuit calculation to determine both the value and direction of the current in each relay for each connection case. The results of load flow analysis and short circuit calculations will be used as inputs to Genetic Algorithm (GA) routine to determine the optimal relays settings to achieve optimal proper coordination between relays at all connection cases. The GA routine will be performed using MATLAB software considering all relays variables in the optimization problem including pick up current (Ip.u), Time Multiplier Setting (TMS), Plug Setting Multiplier (PSM) and characteristic curve type (CS). Also, a novel constraint is added to the optimization problem in order to avoid relays curves intersection for the sake of guaranteeing the prevention of loss of coordination for all fault types and locations. In addition, the proposed algorithm divides the relays in the system into three categories; forward relays, reverse relays, and relays of the connected DGs. This approach guarantees simpler implementation for the optimization problem and good control and satisfaction for all constraints, especially in large Systems. Each relay will send a signal to describe its status (connected or disconnected) to a central unit (central communication). The central unit will identify the connection case of the system. Based on the connection case of the system, the central unit will send a signal to each relay to activate the suitable relay settings as calculated using offline calculations. In the case of High Impedance Faults (HIF), the faulted feeder must be disconnected and isolated in fast, reliable, and accurate way to protect the equipment and people from electrical hazards, especially with the connection of Distributed Generators (DGs). In this context, the detection and isolation of high impedance fault represents a major challenge when it comes to protection engineers as it draws a very low current that conventional overcurrent relays, in most cases, fail to detect. A novel online method, which depends on using Wavelet Transform (WT), will be presented to detect HIF. A symlet Wavelet 8 (symlet 8), the sum of the square of the values of detail 2 (d2) over a cycle, to be used to identify the HIF. The decentralized communication represents the communication between the two relays in each feeder. This communication will increases the accuracy of the decision of relays especially in the case of HIF and also reduces the decision time for all fault cases. The proposed method performance will be tested on IEC benchmark micro grid system. Also, a model simulates the idea of the proposed method will be presented using Real-Time Simulator (RTS) which guarantees accurate and real-time results. In addition, the results of the work will be compared with other works to prove its effectiveness and validation.