A Data-Driven Global Sensitivity Analysis of Output Power to Electrical Faults in Different SPV Array Topologies

The faults in a solar photovoltaic (SPV) array significantly impact the energy balance and system dependability. Lowered system efficiency, reduced output power, and fire risks result from SPV array faults. Evaluating the loss during high impedance array faults is difficult because of the maximum power point tracking (MPPT) control and the blocking and bypass diodes employed in an SPV system. This article analyzes the output power reduction due to different electrical faults in an SPV array connected to a grid. Two different topologies of SPV array configuration, namely, series-parallel (SP) and total-cross-tied (TCT), have been analyzed to find out the sensitivity of power to various electrical faults such as line-to-line (LL) and line-to-ground (LG) at different fault resistances. Using minimal historical data, a data-based global sensitivity analysis using a Kriging-based metamodel approach is proposed to discover the hidden global relationship between output power to various electrical faults in an SPV array. Later, Sobol indices are calculated to assess the severity of faults. The proposed methodology is validated in a hardware set-up of a grid-connected SPV system. The evaluation of performance indicators suggests that SP topology is more efficient as compared to TCT at zero impedance faults, whereas, during high impedance faults, both topologies carry similar power losses.