Power system fault diagnosis based on hybrid rule network and time sequencing information of alarms

Most existing power system fault diagnosis methods employ the state information of protective relays and circuit breakers, while the time sequencing information of alarm messages has not yet been fully used. As the result, definite fault diagnosis results cannot always be achieved especially for complicated and multiple faults with information missing and/or distorting, and sometimes it may result in false fault diagnosis. Under this background, a hybrid rule network based power system fault diagnosis method is developed. First, the largest sets of fault hypotheses are generated by establishing rule networks for various components. Then, a fault hypothesis authentic assessment model employing the time sequencing information of alarms is developed so as to determine the fault components and identify the causes. The developed method can fully use the digital information from sequence of events (SOE) and the analogue information from fault recorders, as well as use the cause-effect relationships and time sequencing information of alarms, so as to effectively improve the fault-tolerance capability of the fault diagnosis method. Finally, a practical fault scenario is served for demonstrating the feasibility and efficiency of the developed fault diagnosis model and method. © 2013 State Grid Electric Power Research Institute Press.