Hardware-in-the-Loop Simulation for Smart Grid Applications
Open Access DepositedAdvancements in Transmission and Distribution Systems
Hardware-in-the-Loop (HIL) simulation plays a pivotal role in advancing modern power systems by enabling real-time, high-fidelity emulation of complex electrical network behaviors. Unlike traditional offline simulations that often involve trade-offs between speed and accuracy, HIL combines the benefits of physical modeling with real-time execution, offering a powerful platform for validating control algorithms and hardware components under realistic operating conditions. These capabilities make HIL particularly valuable for smart grid applications, where precise temporal coordination, accurate waveform reproduction, and rapid system response are essential. This thesis systematically investigates the capabilities and limitations of HIL in both transmission and distribution domains, with the overarching goal of enhancing grid resilience, operational intelligence, and technological readiness for future power infrastructure. To this end, a series of practical HIL-based applications are presented and analyzed. First, the thesis addresses time synchronization challenges in International Electrotechnical Commission (IEC) 61850-based digital substations by developing the first real-time co-simulation framework that integrates heterogeneous time references---namely, a satellite-synchronized Global Positioning System (GPS) clock and a terrestrial-based Precision Time Protocol (PTP) source---across multiple HIL platforms connected via a shared local area network. This framework enables a direct, side-by-side evaluation of timing resilience and supports a cost-effective post-fault reconfiguration algorithm using only current transformer (CT) measurements. The investigation then extends to transmission-level protection, demonstrating for the first time how heterogeneous timing sources influence time-domain relay coordination and fault location accuracy. A comparative study between rate-of-change (ROC) and traveling-wave (TW) methods under varying GPS outage durations reveals the trade-off between sensitivity and tolerance to desynchronization. For power quality analysis, an adaptive harmonic power flow (HPF) algorithm is proposed for hybrid AC/DC transmission networks with voltage source converter (VSC)-based high-voltage direct current (HVDC) links. Unlike conventional fixed-parameter approaches, the algorithm employs a two-layer iterative initialization process that remains stable under varying converter firing angles and extends harmonic analysis up to the 15th order. The method is further stress-tested by incorporating geomagnetic disturbances (GMDs) to assess its applicability under extreme operating conditions. Finally, the thesis develops a cyber-physical digital twin (DT) prototype by coupling HIL with a physical diorama of an urban distribution network. This framework uniquely leverages high-resolution time-series transient waveforms—including voltage, current, and phase angle data—to train a graph-based deep learning model integrating graph convolutional networks (GCNs) and recurrent neural networks (RNNs). A transfer learning mechanism enables rapid adaptation to network topology changes by simply updating the adjacency matrix, eliminating the need for retraining. This represents the first demonstration of a fully integrated HIL–DT platform that combines physical visualization, AI-driven fault diagnosis, and resilience evaluation under both cyber and physical attack scenarios. Collectively, these contributions advance the state-of-the-art in HIL research by bridging physical-layer simulation with data-driven intelligence, offering scalable, adaptive, and high-fidelity solutions for the design, validation, and operation of next-generation smart grids.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.