Establish intelligent fault detection in electrical power system: evolutionary computation in industrial IoT environments

Abstract

A study was conducted on deep transferrable learning techniques for diagnosing faults in building energy frameworks. The research focused on scenarios for cross-operational and cross-system conditions. The Industrial Internet of Things has led to the use of evolutionary computing for fault detection in electrical power systems, which is increasingly important for businesses relying on reliable power systems to maintain operations. The goal of this study was to diagnose the fault in an electrical power system using starling murmuration-optimized Long Short-Term Memory (SMO-LSTM). Datasets from the VSB dataset were collected, and they are arranged as follows: 800,000 observed voltages that are recorded as constants in each of the 8712 samples. 97% accuracy was attained with the suggested approach, SMO-LSTM. In comparison to existing methods, the suggested solution outperforms them in fault detection in electrical power systems.