Scaled Conjugate Gradient-Based Intelligent Microgrid Fault Analysis System

A microgrid fed small three-phase transmission line under various fault conditions is analyzed in this paper with the help of artificial neural network. The transmission line model along with different fault conditions are simulated in MATLAB Simulink. The simulation setup includes three sources, viz., PV solar source, battery, and 3-phase AC source. These sources are followed by inverters, transformer, RC filters, circuit breakers, and a three-phase transmission line model. The three-phase inverter converts DC solar and battery inputs to three-phase AC. The transformer elevates the three-phase AC voltage. RC filters eliminate harmonics generated due to power electronics equipment. The circuit breaker switches between sources so that various fault conditions can be generated with respect to magnitude and time. This fault data is used as an input to train the neural network. The neural network is trained with symmetrical and asymmetrical fault conditions for various duration and time. Some of the samples from generated fault conditions are used to test the neural network. The results are used to comment on accuracy and feasibility of the system.