Electronic Thesis and Dissertation Repository

Thesis Format



Master of Engineering Science


Electrical and Computer Engineering


Grolinger, Katarina


High-Impedance Faults (HIFs) are a hazard to public safety but are difficult to detect because of their low current amplitude and diverse characteristics. Supervised machine learning techniques have shown great success in HIF detection; however, these approaches rely on resource-intensive signal processing techniques and fail in presence of non-HIF disturbances and even for scenarios not included in training data. This thesis leverages unsupervised learning and proposes a Convolutional Autoencoder framework for HIF Detection (CAE-HIFD). In CAE-HIFD, Convolutional Autoencoder learns only from HIF signals by employing cross-correlation; consequently, eliminating the need for diverse non-HIF scenarios in training. Furthermore, this thesis proposes a novel HIF classification approach based on the transformer network stacked with the convolution neural network. To discriminate HIFs from non-fault disturbances, probability distribution-based kurtosis analysis is utilized. The proposed approaches reliably detect HIFs with a 100% success rate in terms of all five metrics of protection system performance, namely accuracy, security, dependability, safety, and sensitivity. The evaluation studies show that proposed approaches outperform the state-of-the-art HIF detection techniques and are robust against noise.

Summary for Lay Audience

A High-Impedance Fault (HIF) typically occurs in a distribution network when a live conductor in a distribution network touches surfaces, such as tree limbs, ground, and gravel. The HIF current is usually too low to cause any direct damage to the power system equipment; however, undetected HIFs are a potential hazard to public safety. The inception of HIFs often causes arcing ignition, which can result in wildfires, life-threatening injuries, and other irreparable damages. Moreover, the undetected HIFs cause grid instability and hinder power system integrity. The traditional protection relays fail to reliably detect the majority of HIFs because of diverse characteristics and the low current magnitude of HIFs. The existing approaches for HIF detection are associated with resource-intensive signal processing techniques and supervised Machine Learning algorithms that are not reliable under diverse non-HIF disturbances. Consequently, this thesis proposes a Convolutional Autoencoder framework for HIF detection (CAE-HIFD), an unsupervised learning-based approach that solely learns from the fault data and omits the requirement of the diverse non-fault scenarios during the training process. Also, this thesis proposes a Transformer-CNN (T-CNN) framework for HIF detection and classification, a deep learning-based approach for reliable identification of fault type in the presence of diverse HIF and non-HIF conditions without requiring resource-intensive signal processing. The non-fault disturbances, such as capacitor and load switching, exhibit characteristics similar to the HIFs; therefore, a probability distribution-based kurtosis analysis is utilized to provide security against non-HIF disturbances. The performance of the proposed approaches is evaluated on the diverse data generated by the IEEE 13-node test feeder, and the response is measure on various challenging case studies. The results show that CAE-HIFD and T-CNN achieve 100\% accuracy and outperformed the state-of-the-art approaches for HIF detection.