Abstract:Microgrids are the most popular power generation technology in recent years due to advancements in power semiconductor technology, but protection is a crucial task when a critical fault occurs in the microgrid. Fault detection under critical conditions is a complex task due to the dissimilar magnitude of the fault current which can be small or large when faulty conditions are changing. In addition to the above circumstances, the presence of distributed energy resources and corresponding uncertainty in the microgrid make the protection task more tricky and critical due to unhealthy weather conditions in cases of renewable energy resources when developing a protection scheme. Existing traditional protection schemes based on threshold-setting principles need modification in such conditions to remove the possibility of misoperation in microgrids. Therefore, in this paper, a deep learning-based protection scheme has been proposed to identify such critical faults as well as provide resiliency against the unpredictable behavior of renewable sources of energy. To explore the behavior of the solar and wind turbine-based sources, uncertainty has been modeled in proposed work. In the protection scheme, mainly four tasks have been performed which are mode identification, fault detection/classification, section identification, as well as location estimation under grid-connected and islanded modes of the microgrid. The overall accuracy of the protection scheme is 98.77% for the mode detection module, while 99.47% and 98.51% for fault detection/classification modules in both modes. To remove any possible hindrance due to communication burden, a free-link communication approach has been considered in the work, which is dependable and accurate.

Fault classification and localization in microgrids: Leveraging discrete wavelet transform and multi-machine learning techniques considering single point measurements

Machine Learning Methods for Fault Diagnosis in AC Microgrids: A Systematic Review

Fault identification, classification, and localization in microgrids using superimposed components and Wigner distribution function

An efficient protection scheme for critical fault detection in microgrid under uncertain scenarios using deep learning algorithm

Intelligent Fault Diagnosis Using Deep Learning for a Microgrid with High Penetration of Renewable Energy Sources

Fault detection and classification in hybrid energy-based multi-area grid-connected microgrid clusters using discrete wavelet transform with deep neural networks

Machine Learning-Based Protection and Fault Identification of 100% Inverter-Based Microgrids

Intelligent Fault Detection Scheme for Microgrids With Wavelet-Based Deep Neural Networks

Universal Adaptive Differential Wavelet Energy-Based Protection Scheme for an AC Microgrid

Detection of High Impedance Faults in Microgrids using Machine Learning

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

Integrating fault detection and classification in microgrids using supervised machine learning considering fault resistance uncertainty

Fault detection and classification of DC microgrid based on VMD

Detection and classification of faults in a microgrid using wavelet neural network

Multi‐agent protection scheme for microgrid using deep learning

A Novel Machine Learning-Based Approach for Fault Detection and Location in Low-Voltage DC Microgrids

Adaptive Protection Combined with Machine Learning for Microgrids

Fault Detection, Classification and Localization Along the Power Grid Line Using Optimized Machine Learning Algorithms

Fault Detection in DC Microgrids Using Short-Time Fourier Transform

Fault Detection and Classification in DC Microgrid Clusters

Assessment of Envelope- and Machine Learning-Based Electrical Fault Type Detection Algorithms for Electrical Distribution Grids