Model-based and Learning Methods for Optimization and Management of Energy Flows in Decentralized Smart Grids

Submission Deadline: 31 December 2026 View: 15 Submit to Special Issue

Guest Editors

Assoc. Prof. Plamen Stanchev

Email: p.stanchev@tu-sofia.bg

Affiliation: Department of Intelligent technology in industry, Faculty of Computer Systems and Technologies, Technical University of Sofia, Sofia, Bulgaria

Homepage:

Research Interests: modern methods for optimization, simulation and artificial intelligence in electrical networks, reliability, energy efficiency in small and medium-sized enterprises, controllers, reinforcement learning, distributed generation, technical and economic assessments of storage systems

Prof. Nikolay Hinov

Email: hinov@tu-sofia.bg

Affiliation: Department of Computer Systems, Faculty of Computer Systems and Technologies, Technical University of Sofia, Sofia, Bulgaria

Homepage:

Research Interests: power electronic converters; analysis, modeling, and control of power electronic devices and systems; systems for decentralized generation of electricity from alternative and renewable energy sources; electric and hybrid cars; smart cities and networks

Summary

The transition toward decentralized smart grids is reshaping the way energy is generated, exchanged, controlled, and consumed. The increasing penetration of distributed energy resources, battery storage systems, electric vehicles, and flexible loads requires advanced methods for accurate modeling, predictive analysis, and real-time decision-making. In this context, model-based approaches and learning-based methods have emerged as powerful tools for optimizing energy flows, improving grid flexibility, enhancing reliability, and supporting resilient operation under uncertainty.

This Special Issue aims to highlight recent advances in mathematical modeling, optimization, artificial intelligence, and data-driven control for decentralized smart grids. It seeks high-quality original research and review papers that address both theoretical developments and practical applications related to energy flow management in distributed and multi-agent environments. Contributions may focus on planning, operation, forecasting, coordination, and control strategies that improve the efficiency, sustainability, and robustness of modern power networks.

Suggested themes include: model predictive control for smart grids; machine learning and reinforcement learning for energy management; optimization of distributed energy resources; microgrid and nanogrid coordination; demand response and load forecasting; battery energy storage scheduling; peer-to-peer energy trading; digital twins for decentralized energy systems; fault diagnosis and predictive maintenance; and uncertainty-aware decision-making in smart grids.

Suggested themes:
· Model-based optimization and control of decentralized smart grids
· Machine learning and reinforcement learning for energy management
· Distributed energy resources coordination in microgrids and nanogrids
· Forecasting, demand response, and flexible load management
· Battery energy storage scheduling and multi-energy system integration
· Digital twins, predictive maintenance, and resilient grid operation

Keywords