A combined adaptive network and fuzzy inference system (ANFIS) approach for overcurrent relay system

Accurate models of overcurrent (OC) relays with inverse time relay characteristics play an important role for the coordination of power system protection schemes. Conventional OC relay modelling using techniques like system identification, parameter estimation, direct data storage and software gave only approximate models. Hence, in this paper a new method for modelling OC relay characteristics curves based on a combined adaptive network and fuzzy inference system (ANFIS) is proposed. In this method, OC relay modeling is done using ANFIS for two types of OC relays (RSA20 and CRP9 with different types and various numbers of membership functions to bring out the optimal design. The simulated results are compared with the published results of analytical method and fuzzy model systems. The results obtained are quite encouraging and will be useful as an effective tool for modelling OC relays.