Detection of Faults in Power System Network to Protect from Cyber Attacks

The electrical power grid is a Cyber Physical System (CPS) that forms the lifeline of modern society. The major concern of the electrical network is safe and reliable operation. A cyber attacker in physical power system operation could cause severe damage to the power grid infrastructure and its reliability by stealthily manipulating SCADA operation. Occurrence of fault in different locations in a power system network would change the magnitude of the fault current. This will make a specific relay to operate and any mis-operation of relay may exist if type and location of fault is not identical properly. This work focuses on fault detection ,its classification and identification of faults on electric power transmission lines using neural network. A feed forward neural network is employed, which is trained with back propagation algorithm. The developed neural network is capable of detecting, classifying and location of various types of faults in a transmission line. Simulation is done using MATLAB to demonstrate that artificial neural network based method is efficient in detecting faults on transmission lines and achieve satisfactory performances. An approach for assessing security during the design phase by neural network is investigated in this work. Experimental results show that by training a back propagation neural network to identify attack patterns, possible attacks can be identified from design scenarios presented to it. The performance of the neural network for various types of faults is presented in this work. Neural network is implemented in arduino and fault is identified using LEDs.