Fault section estimation in power systems by discrimination analysis

AbstractOnce a fault occurs in a power system, the fault section must be estimated from inadequate information on protective relays and circuit breakers. There have been several studies to develop an automatic estimation method, e.g., knowledge engineering, neural network, etc. However, it was very difficult to obtain a method which is powerful enough and easy to maintain.In this paper, the discrimination analysis theory is applied to the fault section estimation. Using the theory, it is expected that the inference speed can be improved. It is expected also that the knowledge about the fault section estimation can be collected.First, using the small model system, the fault section estimation method with the discrimination analysis theory is described and compared to the method with neural network. Next, it is illustrated that the method can be applied to a large system.