A novel neural network and backtracking based protection coordination scheme for distribution system with distributed generation

With the increased installation of renewable energy based distributed generations (DGs) in distribution systems, it brings about a change in the fault current level of the system and causes many problems in the current protection system. Hence, effective protection schemes are required to ensure safe and selective protection relay coordination in the power distribution system with DG units. In this paper, a novel adaptive protection scheme is proposed by integrating fault location with protection relay coordination strategies. An automated fault location method is developed using a two stage radial basis function neural network (RBFNN) in which the first RBFNN determines the fault distance from each source while the second RBFNN identifies the exact faulty line. After identifying the exact faulty line, then protection relay coordination is implemented. A new protection coordination strategy using the backtracking algorithm is proposed in which it considers the main protection algorithm to coordinate the operating states of relays so as to isolate the faulty line. Then a backup protection algorithm is considered to complete the protection coordination scheme for isolating the malfunction relays of the main protection system. Several case studies have been used to validate the accuracy of the proposed adaptive protection schemes. The results illustrate that the adaptive protection scheme is able to accurately identify faulty line and coordinate the relays in a power distribution system with DG units. The developed adaptive protection scheme is useful for assisting power engineers in performing service restoration quickly so as to decrease the total down time during faults.