Protection framework for microgrids with inverter‐based DGs: A superimposed component and waveform similarity‐based fault detection and classification scheme

Abstract

Since the penetration level of the inverter‐based distributed generations (IBDGs) into microgrids (MGs) is increasing, the protection issues of such networks have become more challenging. The present work aims to design a protection framework for fault detection and classification in IBDG dominated MGs. In the proposed approach, the current waveforms measured from one end of the protected lines are processed using superimposed components. Then, the faults are identified using the Euclidean distance concept. A Pearson correlation coefficient‐based approach is also developed to characterize the faulty phases in the proposed fault classification unit. Furthermore, an auxiliary index is defined to discriminate load change conditions from the faults. In addition to being simple and efficient, the proposed technique is highly capable of functioning in different MG operating modes and configurations without changing its settings. Also, it demonstrates robust performance against IBDG impacts, noises, and stimulating fault scenarios. To assess the scheme proficiency, a modified IEEE 15‐bus system with highly penetrated IBDGs is simulated. The results confirm the quickness and high accuracy of the proposed technique. Moreover, the method is validated in an experimental laboratory test bench. Finally, the superiority of the method is confirmed by a comparative study with other recent methods.

Hardware Supported Fault Detection and Localization Method for AC Microgrids Using Mathematical Morphology with State Observer Algorithm

“…Compared benchmark Methods Proposed method Harmonic content-based method [17] State observerbased method [24] Kalman filterbased method [27] CNN-based method [ The proposed scheme successfully detected the fault in less than 15 milliseconds in almost all cases. Therefore, by comparison, it is observed that this detection time is much less than most of the existing methods.…”

“…Authors in [16] utilized the online current wavelet injection method, traveling waves, and mathematical morphology to protect microgrids from different kinds of faults. Euclidean distance was deployed in [17] to compute superimposed components with waveform similarity for the fault's detection/classification and localization. A Time-timetransform was proposed in [18] for microgrid protection in both the OGM and IM operations.…”

DC line fault discrimination based on maximum information coefficient

“…Using this relationship, it is of practical significance to mine the hidden relationships between big data. The reciprocal information of the currents at different sides of the line has a significant difference at the time of the fault, providing a viable path for fault discrimination within and outside the zone using the reciprocal information between the currents at both ends [4][5] .…”

FPGA-based fault analysis for 7-level switched ladder multi-level inverter using decision tree algorithm

“…The line location and variable impedance faults happening in the transmission line connected inverter-generator can be diagnosed using the machine learning-based technique [13]. The faulty phases are extracted by the Pearson correlation coefficient-based technique for the micro-grid fed inverter based generator [14]. The classification and regression tree (CART) method can be utilized for the fault location of a single-phase grounding in real-time power converters [15].…”