A Comprehensive Review of Fault Detection, Classification, and Location Techniques in Transmission Networks for Developing Countries
Ukut, Uwem, Isong
Department of Electrical and Electronic Engineering, University of Uyo, Uyo, Nigeria.
Nsebong Opura
Department of Electrical and Electronic Engineering, University of Uyo, Uyo, Nigeria.
Akaninyene Obot
Department of Electrical and Electronic Engineering, University of Uyo, Uyo, Nigeria.
Udofia, Kufre *
Department of Electrical and Electronic Engineering, University of Uyo, Uyo, Nigeria.
K. Akpabio, Itoro
Department of Electrical and Electronic Engineering, University of Uyo, Uyo, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
Reliable transmission networks underpin national economic development, yet developing countries continue to experience disproportionately high rates of transmission line faults, prolonged outage durations, and constrained investment in protection infrastructure. This review synthesises the current state of knowledge on fault detection, classification, and location techniques applicable to high-voltage transmission systems, with particular attention to the technical, economic, and institutional constraints that shape technology adoption in low- and middle-income power sectors. Conventional protection philosophies based on impedance relaying and overcurrent schemes are examined alongside signal-processing approaches such as wavelet transforms and travelling-wave methods, and against the growing body of work applying machine learning and deep learning architectures, including convolutional neural networks, long short-term memory networks, and hybrid ensembles, to fault diagnosis tasks. The review finds that although artificial-intelligence-based methods report consistently high accuracy under simulated conditions, their transferability to developing-country networks is constrained by sparse instrumentation, weak communication infrastructure, limited synchrophasor coverage, and a shortage of locally labelled fault data. High-impedance faults, series compensation, renewable-integrated feeders, and ageing conductor assets introduce further complications that are underrepresented in the literature, which remains dominated by simulation studies from well-instrumented grids. The review identifies practical pathways for closing this gap, including low-cost phasor measurement architectures, transfer learning from synthetic to field data, and hybrid schemes that combine physics-based fault location with data-driven classification. The synthesis is intended to orient researchers, utility engineers, and regulators in developing economies towards protection strategies that are both technically sound and economically deployable within prevailing infrastructure constraints.
Keywords: Transmission line protection, fault detection, fault classification, fault location, developing countries, machine learning, power system reliability.