A Communication-Free Fault Detection and Classification Scheme for Power Islands

Fault detection and classification (FDAC) for islanded power networks is a challenge due to low short circuit current fed by inverter interfaced distributed generator (IIDGs), control dependent fault models and bi-directional current flow. This article proposes a FDAC scheme, which utilizes the local voltages, measured at relay terminals. The transient monitoring function (TMF) of the voltage signal ($T_{v}$) gives the information of the fault transients. The proposed scheme is validated on the modified CIGRE test microgrid and IEEE 33 bus distribution network. Several test cases are simulated in MATLAB/Simulink at various locations with different inception angles, loading conditions, and fault resistances. The obtained results proves the accuracy of proposed FDAC scheme, with local information available at relaying points. The relays can be programmed to adapt to the proposed scheme, and hence there is no need for additional communication. Experimental validation with Texas Instruments digital signal controller (DSC) TMS320F28377D and OPAL-RT real time digital simulator shows the efficacy of the proposed scheme for practical implementation. Further, performance during no-fault scenarios, and accuracy with different controllers and current limiting algorithms are also illustrated.