Digital Twin–Based Predictive Fault Detection and Performance Optimization for Hybrid HVDC–FACTS Networks

Abstract

The increasing penetration of high-voltage direct current (HVDC) transmission and flexible AC transmission system (FACTS) technologies has significantly enhanced the controllability and efficiency of modern power grids, while simultaneously increasing system complexity and operational vulnerability. This research proposes a Digital Twin–based predictive fault detection and performance optimization framework for hybrid HVDC–FACTS networks, aimed at enabling real-time monitoring, early fault diagnosis, and intelligent operational decision-making. The proposed digital twin integrates high-fidelity physical models of HVDC converters and FACTS devices with real-time measurement data to achieve continuous synchronization between the physical system and its virtual counterpart. Online parameter estimation techniques are employed to track dynamic variations in system components, enabling early detection of incipient faults and performance degradation before critical failures occur. Artificial intelligence–assisted analytics, including data-driven anomaly detection and predictive maintenance models, are incorporated to enhance fault classification accuracy and remaining useful life (RUL) estimation. The framework further supports adaptive control and performance optimization by dynamically tuning operating parameters to maintain system stability, improve power quality, and minimize energy losses under varying operating conditions. Results indicate that the proposed digital twin framework can achieve high-fidelity real-time system representation with synchronization errors below predefined operational thresholds, enabling fault detection significantly earlier than conventional protection and monitoring schemes. Simulation-based validation is anticipated to demonstrate improved fault prediction accuracy, reduced unplanned outages, enhanced converter reliability, and measurable improvements in system efficiency and voltage stability. Overall, the study is expected to confirm that digital twin–enabled predictive intelligence offers a scalable and industry-ready solution for enhancing the resilience, reliability, and operational efficiency of future hybrid HVDC–FACTS power networks.