Power, Control, and Data Processing Systems

Abstract Reinforcement learning has emerged as a valuable tool in control theory and related areas, such as robotics and process control, facilitating prediction, identification, and management of complex systems. A key advantage of using reinforcement learning is its ability to derive optimal control policies without requiring a comprehensive understanding of the underlying system dynamics. In essence, RL develops control strategies through interaction with the environment. A common approach in controller design for these systems involves the use of approximation methods. Given the complexities associated with solving PDEs analytically, numerical techniques like the finite element method are typically employed for approximation. In this research, we begin by discretizing the PDE that characterizes the system's dynamics using an appropriate discretization technique. Following this step, we extract the discrete dynamics of the system. Since these discrete dynamics exhibit the Markov property where the future state depends only on the current state and the action taken the next phase involves designing a controller based on these derived dynamics. Recognizing that reinforcement learning focuses on optimizing future actions through data analysis, feedback from the optimal mode can be utilized as a viable option for controller design. This study will further explore LQR controller design for one-dimensional heat equation heat flow as a system whose dynamic is described with PDE.

Abstract The aim of this study is to examine how the integration of two innovative technologies Artificial Intelligence (AI) and Blockchain can transform and enhance the efficiency, security, and transparency of Iran’s electricity sector. Accordingly, this research adopts a descriptive–analytical approach based on a systematic review of national and international academic sources to identify the capacities, challenges, and strategies for integrating these technologies into the country’s energy system. The findings indicate that AI, through predictive analytics, adaptive learning, and process optimization, can make electricity network management smarter and more efficient. Conversely, blockchain technology provides a transparent, secure, and decentralized platform for energy data exchange. The synergy between these two technologies can lead to reduced energy losses, facilitation of peer-to-peer energy trading, enhanced data security, and improved grid sustainability. However, several barriers hinder this integration, including scalability limitations, high implementation costs, incompatibility with legacy infrastructure, legal and regulatory challenges, and a shortage of skilled professionals. Finally, the study proposes strategic measures such as developing supportive policies, employing deep learning–based smart contracts, adopting low-energy blockchain solutions, and expanding IoT-based energy management systems. These actions can pave the way for the establishment of an intelligent, efficient, and sustainable electricity infrastructure in Iran.

Abstract Smart grid and demand side management have increased the interaction between energy suppliers and consumers. Energy savings are possible through optimal resource allocation and energy utilization. This interaction helps reduce grid dependency, optimize resources. This paper reviews the socio-economic and reliability benefits of smart grid and focuses on the classification and analysis of recent studies that propose game theory approaches for optimizing demand side management. The analysis of each application, including cooperative and non-cooperative games, is carried out by analyzing the mechanisms and solution methods. Based on the review of these studies, game theory applications on the demand side can be beneficial for the grid and the consumer. Classification and review of recent works, proposing game theory approaches for optimal demand side management are considered in this paper. Applications of different cooperative and non-cooperative games are analyzed by reviewing their mechanisms and solution methods. It is concluded from the review of these works that game theory applications on demand side is beneficial for both grid and the consumer.

Abstract This study introduces a hybrid framework that integrates supervised machine learning (ML) algorithms with the VIKOR (VlseKriterijumska Optimizacija I Kompromisno Resenje) multi-criteria decision-making (MCDM) technique to advance anomaly detection and cell ranking in next-generation networks. The proposed model addresses critical challenges in heterogeneous network environments, including data imbalance and fault prioritization. Three ML algorithms—Naïve Bayes, Decision Tree, and Random Forest—were evaluated, with Random Forest achieving the highest accuracy (93.658%). However, the Decision Tree algorithm demonstrated optimal efficiency, balancing high accuracy (92.688%) with the fastest execution time (0.04 seconds), rendering it particularly suitable for real-time applications. The incorporation of VIKOR enhanced the framework by enabling fault prioritization based on severity and impact, improving detection of minority fault classes, and supporting multi-criteria resource management. This hybrid approach resulted in improved system accuracy, flexibility, and scalability, ultimately contributing to reduced operational response times and enhanced network reliability. The findings validate the efficacy of combining ML with MCDM for intelligent fault management and cell ranking in complex network ecosystems

Abstract One of the important topics in the use of Faraday cup in plasma focus devices is the production of secondary electrons due to the collision of beam particles with the inner surface of the Faraday cup and their trapping in the measurement of particle beam flow by the cup. Therefore, in this study, the simulation of the geometrical effects such as the depth and Internal structure of the cup on the production of secondary electrons and the output current of the Faraday cup has been done. The used simulation tool is CST STUDIO SUITE. Based on the simulation results, reducing the aspect ratio leads to an increase in the particle trapping ability, and as a result, the flow measured by the Faraday cup is closer to the real flow. Furthermore, The Faraday cup current for a cylindrical sample is greater than the actual current and in the ramp example, the Faraday cup current is closer to the actual value.

Abstract Electric vehicles and hydrogen and gas storage systems, provide significant flexibility for energy management in the smart grid. This paper extends the integrated management framework of renewable energy resources, demand response programs, and water systems by incorporating two advanced electricity storage technologies: Power-to-Gas (P2G) and Compressed Air Energy Storage (CAES). Unlike conventional applications, in this work both units are modeled solely as electrical storage layers to enhance flexibility and enable price arbitrage under fixed, Time-of-Use (TOU), and real-time pricing schemes. Surplus renewable generation can be stored either as synthetic gas in the P2G system or as compressed air in the CAES unit, both of which are later converted back into electricity during high-price periods. The proposed extended model improves operational efficiency without introducing additional heat or gas loads. Simulation results show that integrating P2G and CAES reduces total operational cost by 4.5%, increases renewable utilization by 10%, and enhances the ability to shift consumption away from expensive hours.