Adaptive Current Differential Protection Method Based on Fréchet Distance Algorithm

The integration of a large number of distributed generation (DG) units has altered the grid structure and fault characteristics of distribution networks, posing significant challenges to conventional protection methods. To address this, this paper proposes an adaptive current differential protection method based on comparing the similarity of current waveforms at both ends of a line. After a fault occurs, the current waveforms at both ends of each line are first extracted and normalized. The Fréchet distance algorithm is then introduced to quantify the waveform similarity. Based on the calculated Fréchet distance, the restraint coefficient of the current differential protection is constructed and the protection criterion is improved. Finally, a logic function is derived from the improved criterion. A short-circuit fault within the section is identified when the sum of the logic functions across all sampling points exceeds 50% of the total number of sampling points. A simulation model built in PSCAD/EMTDC is used for validation. Simulation results demonstrate that the proposed method is unaffected by fault type, transition resistance, fault location, or DG grid-connected capacity, has low data synchronization requirements, and exhibits excellent reliability and selectivity.