This study describes a support vector machine (SVM)-based technique for identifying loss-of-excitation (LOE) condition in synchronous generators from other disturbances such as external faults and power-swing conditions. In this new approach, only one zone of LOE is required and the time coordination is reduced significantly. The proposed method is compared with traditional two-zone impedance method. Several operating conditions within the generator capability are used to verify the generality of the SVM-based classifier. The proposed classifier identifies an LOE condition in all cases before the impedance enters the larger mho impedance zone. Faults and power-swing conditions are identified correctly, thereby preventing incorrect operation of the LOE impedance zone.
A power transformer is one of the most important and expensive components in any power system. Power transformers can be exposed to a wide variety of abnormal conditions and faults.Internal turn-to-turn faults are the most difficult types of faults to detect within the power transformer. The IEEE Standards documents have revealed that there is no one standard way to protect all power transformers against minor internal faults such as turn-to-turn faults and at the same time to satisfy basic protection requirements: sensitivity, selectivity, and speed.This thesis presents a new, simple and efficient protection technique which is based on negative sequence currents. Using this protection technique, it is possible to detect minor internal turn-toturn faults in power transformers. Also, it can differentiate between internal and external faults.The discrimination is achieved by comparing the phase shift between two phasors of total negative sequence current. The new protection technique is being studied via an extensive simulation study using PSCAD®/EMTDC™ 1 software in a three-phase power system and is also being compared with a traditional differential algorithm.Relay performance under different numbers of shorted turns of the power transformer, different connections of the transformer, different values of the fault resistances, and different values of the system parameters was investigated. The results indicate that the new technique can provide a fast and sensitive approach for identifying minor internal turn-to-turn faults in power transformers.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.