System Modeling-Based Verification of Datasets for Reliable Data-Driven Digital Twin Development in Grid-Connected Solar Photovoltaic Inverter
IEEE Transactions on Reliability · 2026 · 75 · 2145-2158
DOI: 10.1109/tr.2026.3701974 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Reliability, 2026. Verfasst von Arun Kumar und 1 weiteren Autor:innen. Die Arbeit nennt 23 Literaturangaben. Gefördert von: Government of India through the Anusandhan National Research Foundation, Intelligent Digital Twin Framework for Solar and Wind-Based Microgrid: Real-Time Health Monitoring.
Abstract (Originalsprache)
Ensuring reliable operation of grid-tied photovoltaic (PV) inverters is critical as semiconductor aging, environmental variability, and grid disturbances pose major threats to system stability. This article presents a data-driven digital twin (DD-DT) framework for real-time predictive monitoring of a two-stage single-phase grid-connected PV inverter, incorporating physics-aware statistical cross-validation (PASCV) and temporal convolutional networks (TCN). The proposed framework addresses critical challenges such as sensor noise, measurement faults, and system variations by integrating a physics-based validation layer with DD learning. PASCV ensures only trusted measurements are used for model training and real-time synchronization, while TCN enables high-fidelity time-domain prediction of key inverter states under dynamic operating conditions. Experimental validation is carried out on an NIsbRIO-OPAL-RT-based hardware testbed under diverse scenarios, including datasets with 2, 3, and 4 corrupted signals with irradiance variations and grid voltage disturbances. Results demonstrate that the DD-DT achieves superior predictive accuracy (mean absolute error ∼ 3.40%, mean absolute percentage error ∼3.17%, and correlation coefficient ≥ 0.971) compared to conventional DT approaches. The proposed reliability-centric DD-DT thus provides a scalable pathway toward reliable DTs for power electronic circuits in renewable energy systems.
Förderung
- Government of India through the Anusandhan National Research Foundation
- Intelligent Digital Twin Framework for Solar and Wind-Based Microgrid: Real-Time Health Monitoring
Zitieren
APA
Kumar, A., & Kumar, N. (2026). System Modeling-Based Verification of Datasets for Reliable Data-Driven Digital Twin Development in Grid-Connected Solar Photovoltaic Inverter. IEEE Transactions on Reliability, 75, 2145-2158. https://doi.org/10.1109/tr.2026.3701974
Vancouver
Kumar A, Kumar N. System Modeling-Based Verification of Datasets for Reliable Data-Driven Digital Twin Development in Grid-Connected Solar Photovoltaic Inverter. IEEE Transactions on Reliability. 2026;75:2145-2158. doi:10.1109/tr.2026.3701974
BibTeX
@article{Kumar2026,
author = {Kumar, Arun and Kumar, Nishant},
title = {System Modeling-Based Verification of Datasets for Reliable Data-Driven Digital Twin Development in Grid-Connected Solar Photovoltaic Inverter},
journal = {IEEE Transactions on Reliability},
year = {2026},
volume = {75},
pages = {2145--2158},
doi = {10.1109/tr.2026.3701974}
}
EUVIMED