Fault Detection and Classification in Parallel and semi-parallel Transmission Lines Connected to Large-Scale Photovoltaic Power Plants Using Artificial Neural Networks and Support Vector Machines

Sustainable energy supply and the increasing use of large-scale photovoltaic power plants (LSPPPs) demand stronger and more reliable transmission systems. Parallel transmission lines (PTL) are extensively utilized in contemporary power systems to augment the reliability and security of transmission line. With the need for energy transfer from LSPPPs and the impact of LSPPPs on changing the nature of faults, PTLs face new challenges. This paper investigates the fault detection (FD) and fault classification (FC) with various characteristics in parallel and semi-parallel transmission lines using Artificial Neural Networks (ANNs) and Support Vector Machines (SWM). This approach facilitates the enhancement of efficiency and confidence in transmission systems and contributes to organizing stable and reliable performance in these systems. For simulation and analysis, the PSCAD/EMTDC software is employed to simulate the model of parallel and semi-parallel transmission lines connected to LSPPPs. Subsequently, the three-phase current signal values and the positive sequence voltage angle signals of the two buses connected to the transmission line are extracted, processed, and divided into training and testing data using MATLAB software. Then, these two methods have been compared. The results demonstrate that ANN and SWM are capable of accurately FD and FC with high precision.