AI Summary & Key Takeaways
AI InsightSimulation results demonstrate the NN fault location estimation in percent of line length are in a suitable range.
📌 3 Key Takeaways
Abstract
Nowadays, microgrids have attracted much attention in developed countries. The protection of DC systems, unlike conventional AC systems, is a highly challenging task. The acquaintance on fault location in distribution network causes for quick restoration, maintenance and decrease unnecessary power outage period. Neural Networks (NNs) are among the powerful, reliable approaches and are used in many different engineering applications. Also, Multi-Layer Perceptron (MLP) NNs are used for different estimating problems. This paper presents an accurate protection method for Low-Voltage DC (LVDC) ring-bus microgrid systems based on MLP NN. The aim of the proposed method is precise fault location estimation in microgrids, irrespective of the type and magnitude of fault, current, and the power supply quantity, by instantaneous current monitoring of each segment of the microgrid. Simulation results demonstrate the NN fault location estimation in percent of line length are in a suitable range. The results show that the estimation error is small and is within the permissible range. According to the results, efficiency and accuracy of MLP NN are confirmed. To do so, an LVDC ring-bus microgrid is used that utilizes solid-state bidirectional switches along with master and slave controllers