Fault Detection and Location Method for Mesh-Type DC Microgrid Using Pearson Correlation Coefficient

Fault detection and location method are two significant techniques to recognize and clear the fault timely. This paper proposes a novel fault detection and location method for mesh-type DC microgrid. The key of this method is to implement similarity analysis on the sampled lines current and the lines current reference by employing an improved Pearson correlation coefficient. Firstly, the fault detection can be implemented by comparing the sampled lines current with steady-state lines current in a movable time window. And the fault line and fault type can be recognized based on the analysis on the correlation coefficient. Secondly, the fault location can be obtained by comparing the sampled lines current with calculation results of transient lines current, which corresponds to the estimated fault location scheme, within a fixed time window. The estimated location scheme, including the fault position and fault resistance, is updated by genetic algorithm, and will be determined when the correlation coefficient is higher than the threshold. Moreover, both numerical simulation and experiment results validate that the proposed method can recognize the fault line and fault type timely, while obtain the fault position and fault resistance accurately.

Automated summary

A novel fault detection and location method for mesh-type DC microgrid is proposed in this paper, which utilizes similarity analysis and correlation coefficient analysis to achieve fault detection and location, with the updated fault location scheme by genetic algorithm.