Optimal Coordination of the Overcurrent Relay Protection of Power Systems

• Main Authors: Cheng-Hung Lee, 李振弘
• Other Authors: Chao-Rong Chen
• Format: Others
• Language: en_US
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/yvzrjk

博士 === 國立臺北科技大學 === 電機工程系博士班 === 101 === The dissertation is to research the optimal coordination of the overcurrent relay (OCR) protection of power systems for adaptive protections. OCR has long been the main device protecting power systems. These devices were originally based on electromechanical induction, and new device types were then developed based on solid-state, microprocessors and intelligent electronic devices (IED), with the protective features evolving from single phase units to multi-phase multi-functional units with communications and network SCADA control. Overcurrent relay with a single fundamental structure is cost-effective and selective, but the relay settings for each protective application should be carefully ranked and upstream to downstream protective isolation coordination should be considered to achieve adaptive protection. The research starting from introduction of the OCR fundamental theory, construction and development to the coordination works. The aim of this study is to investigate the optimal methodologies that can be used to locate the optimal settings of each relay in various power systems. Genetic algorithms and immune algorithms are adapted for application to protective relay coordination, and their performance is compared with the conventional method and with each other. The blind spots of optimal methods is discussed, a partial differentiation approach (PDA) is introduced to solve the problem, and the concept of the state-of-the-art application of the IED is discussed for the case of an adaptive load change. The results show that the proposed methods are able to find the optimal settings of overcurrent relays not only for radial type but also for complicated interconnected power systems, an important finding as it is often difficult to manually set and verify the complicated systems.