A Novel Machine Learning-Based Framework for Optimal and Secure Operation of Static VAR Compensators in EAFs

A static VAR compensator (SVC) is a critical component for reactive power compensation in electric arc furnaces (EAFs) that used to relieve the flicker impacts and maintain voltage level. weak profile can not only reduce power-quality services, but also result system instability severe cases. The cybersecurity of EAFs becoming significant concern due their cyber-physical structure. reliance SVC controllers on measurement network communications has resulted cyber-vulnerability point unauthorized access EAF, which affect its normal operation. This paper addresses concerns about cyber attacks EAFs, cause communication issues data SVCs. Three different types are launched controllers—a replay attack, delay false injection attack (FDIA)—were simulated investigated. In order stop activities attacks, secured anomaly detection model (ADM) based prediction interval proposed. proposed dependent support vector regression new smooth cost function constructing optimal symmetrical intervals. modified algorithm teaching–learning-based optimization was developed adapt ADM’s parameters during training. simulation’s outcomes genuine dataset showed strong capability against EAFs.