Insights and Recommendations for the Assessment and Design of Fault Diagnosis Methods Applied to Modern Distribution Systems

The growing integration of Distributed Generation (DG) and the complexity of modern distribution networks present new challenges in fault diagnosis. Thus, this paper comprehensively reviews fault diagnosis methods in distribution systems, emphasizing modeling aspects for real-world applicability. In addition, this study critically evaluates the tests performed when proposing methodologies for fault detection, classification, and location in both overhead and underground networks, highlighting limitations such as the lack of consideration for variations in operating conditions and DG diversity. Separate analyses are performed for overhead lines and underground networks, which are not easily found in the literature. A bibliometric analysis was performed to identify research trends, revealing a focus on signal processing techniques like Wavelet Transform and the increasing use of machine learning methods. The analysis also emphasizes the importance of robust distribution systems modeling, which includes accurate measurement systems and communication network. The findings aid future research to improve the reliability and efficiency of fault diagnosis systems in distribution networks, address current gaps, and point the way for more practical and adaptable solutions.