Estimate, Simulate, and Optimize: Approaches to Voltage Control in Smart Grids with On-Load Tap Changer Transformer and Advanced Metering Infrastructure

• Autores: Iker García Ribote
• Directores de la Tesis: Roberto Santana Hermida (dir. tes.)
• Lectura: En la Universidad del País Vasco - Euskal Herriko Unibertsitatea ( España ) en 2025
• Idioma: inglés
• Programa de doctorado: Programa de Doctorado en Química Teórica y Modelización Computacional/Theoretical Chemistry and Computational Modelling por la Universidad Autónoma de Madrid; la Universidad Complutense de Madrid; la Universidad de Barcelona; la Universidad de Cantabria; la Universidad de Extremadura; la Universidad de las Illes Balears; la Universidad de Murcia; la Universidad de Oviedo; la Universidad de Sevilla; la Universidad de Vigo; la Universidad del País Vasco/Euskal Herriko Unibertsitatea; la Universida
• Calificación: Sobresaliente Cum Laude
• Enlaces
  • Tesis en acceso abierto en: ADDI
• Resumen

  • This dissertation explores the evolving challenges and advancements in voltage regulation within the field of electrical power distribution, particularly emphasizing the integration of smart grid technologies. The research is structured into three pivotal chapters, each addressing a distinct aspect of voltage regulation through innovative machine learning and optimization techniques. The initial chapter introduces a groundbreaking modeling framework for on-load tap-changer transformers, which are crucial for real-time voltage regulation in smart grids. This framework not only evaluates various control strategies but also proposes a comprehensive evaluation system, thereby enhancing the effectiveness and efficiency of these transformers. The second chapter addresses the challenges of real-time data handling in smart grids by introducing a new estimation problem known as the Unattributed-Identity Multi-Target Regression problem. A tailored feature selection methodology is developed, significantly improving predictive accuracy and efficiency in handling data from IoT devices, thus enhancing grid observability and operational efficiency. The final chapter enhances the accuracy of electric distribution companies' topology mapping through the development of the Distance-based Neighborhood k-Search method. This method outperforms existing techniques in detecting and correcting virtual connectivity models, thereby ensuring smoother and more reliable grid operations. This research not only advances the theoretical frameworks and methodologies in smart grid management but also provides robust, industry-validated solutions that address the complexities of modern electrical networks. The collaboration with leading energy sector companies underscores the practical relevance and applicability of the findings across different geographical markets, setting a new benchmark for future research in smart grid voltage regulation.