Hierarchical fault diagnosis for power systems based on equivalent-input-disturbance approach

This paper describes a new fault diagnosis method for power systems that is based on a hierarchical model and the equivalent-input-disturbance (EID) approach. First, a hierarchical model describing the system is built. Next, an EID estimator is designed for each submodel in each layer. Then, the estimators are used to carry out fault diagnosis from the top down. Unlike conventional methods of fault diagnosis, which generally involve the fusion of extensive sensor information from protective relays and circuit breakers, this method requires a very small amount of information on the system to rapidly perform fault diagnosis. Simulations of the IEEE 13-bus radial distribution feeder and a comparison with other methods demonstrate the validity and superiority of the method.