Examinando por Autor "Hidalgo, Mauricio"

Mostrando 1 - 2 de 2
  • Miniatura
    Ítem
    A Digital-Twin-Enabled Resilience Framework (DTERF) for Machine-Learning-Based Anomaly Detection in High-PV Cyber–Physical Smart Grids
    (MPDI, 2026-08-26) Yanine, Franco Fernando; Hidalgo, Mauricio; Frez, Jonathan; Rao, Challa Krishna; Sahoo, Sarat Kumar
    The rapid integration of solar photovoltaic (PV) generation, distributed energy resources, and advanced communication infrastructures is transforming conventional power systems into highly interconnected cyber–physical smart grids. Although this transition improves sustainability and operational flexibility, it also increases grid-management complexity and introduces cyber–physical vulnerabilities, including false data injection attacks, communi￾cation failures, equipment degradation, and renewable-induced operational instabilities. This paper presents the Digital-Twin-Enabled Resilience Framework (DTERF), a conceptual reference architecture for anomaly detection in high-PV cyber–physical smart grids. DTERF integrates heterogeneous cyber–physical data acquisition, Digital Twin-based contextual representation, machine-learning analytics, explainable decision support, adaptive opera￾tional response, continuous learning, and self-healing capabilities within a unified resilience cycle. The framework is grounded in a structured review and comparative assessment ofcontemporary machine-learning approaches and recent integrated smart-grid research. Itsarchitecture is conceptually evaluated through requirements-to-architecture traceability, examining functional coverage and internal consistency across the complete operational cycle. The analysis shows that DTERF provides explicit architectural mechanisms address￾ing the principal requirements identified in the literature, including contextual anomaly analysis, interpretability, cybersecurity robustness, resilience support, and operational inte￾gration. Rather than proposing a new anomaly detection algorithm or claiming empirical performance superiority, DTERF provides a technology-agnostic architectural foundationfor coordinating complementary capabilities required for resilient anomaly management. Future work should empirically validate the framework using Digital Twin simulation envi￾ronments, representative high-PV distribution systems, cyber–physical anomaly scenarios, and real or utility-derived operational data.
  • Miniatura
    Ítem
    Limitations of transfer learning for chilean cherry tree health monitoring: when lab results do not translate to the orchard
    (MDPI, 2025-08-13) Hidalgo, Mauricio
    Chile, which accounts for 27% of global cherry exports (USD 2.26 billion annually), faces a critical industry challenge in crop health monitoring. While automated sensors monitor environmental variables, phytosanitary diagnosis still relies on manual visual inspection, leading to detection errors and delays. Given this reality and the growing use of AI models in agriculture, our study quantifies the theory–practice gap through comparative evaluation of three transfer learning architectures (namely, VGG16, ResNet50, and EfficientNetB0) for automated disease identification in cherry leaves under both controlled and real-world orchard conditions. Our analysis reveals that excellent laboratory performance does not guarantee operational effectiveness: while two of the three models exceeded 97% controlled validation accuracy, their field performance degraded significantly, reaching only 52% in the best-case scenario (ResNet50). These findings identify a major risk in agricultural transfer learning applications: strong laboratory performance does not ensure real-world effectiveness, creating unwarranted confidence in model performance under real conditions that may compromise crop health management.