Research Articles
Power systems
Hamid Reza Sezavar; Hamid Karimi; Navid Fahimi
Abstract
A comparative approach is pretend in this paper that evaluates different Artificial Intelligence (AI) methods for diagnosing power transformer faults using Dissolved Gas Analysis (DGA). Traditional approaches like the Rogers Ratio Method and Duval Triangle have been used for many years, but offer unreliable ...
Read More
A comparative approach is pretend in this paper that evaluates different Artificial Intelligence (AI) methods for diagnosing power transformer faults using Dissolved Gas Analysis (DGA). Traditional approaches like the Rogers Ratio Method and Duval Triangle have been used for many years, but offer unreliable results for complex cases. Although, newer AI methods present better results, but still vary in how well they work. In this paper, several AI approaches are evaluated including Support Vector Machines (SVMs), Random Forest (RF), Gradient Boosting Machines (GBMs), Deep Neural Networks (DNNs) and a new combinational model is proposed based on comparing results. A real DGA dataset is used for covering six different fault types for the proposed testing. The results show that while all AI methods do better than traditional approaches, the combinational approach performs the best with 92.3% accuracy. This is found 20.2% better than traditional methods and 4.8% better than the best single AI model. Rational explanation is provided for how each method works and practical recommendation is presented for choosing the right approach based on particular requirements and available resources in real practices.
Research Articles
Computer science
mahmodreza hajian; Ali Broumandnia; afshin salajegheh; Razieh Farazkish
Abstract
The rapid expansion of the Internet of Things (IoT) has raised critical security concerns across its heterogeneous and resource-constrained networks. Ensuring robust protection with minimal overhead requires identifying a minimal set of key nodes that can provide wide security coverage. This study introduces ...
Read More
The rapid expansion of the Internet of Things (IoT) has raised critical security concerns across its heterogeneous and resource-constrained networks. Ensuring robust protection with minimal overhead requires identifying a minimal set of key nodes that can provide wide security coverage. This study introduces a novel framework that integrates graph-theoretic modeling with metaheuristic optimization to enhance IoT network security. The proposed approach formulates the problem as an External Independent Dominating Set (EIDS), where selected security nodes ensure full coverage while maintaining independence to reduce correlated vulnerabilities. Four metaheuristic algorithms—Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Bee Colony Optimization (BCO), and Simulated Annealing (SA)—are implemented and compared using multiple network topologies, including random graphs, wireless sensor networks, and real-world smart city infrastructures. Experimental results show that SA achieves the best coverage-to-cost ratio, while PSO offers superior computational efficiency. BCO demonstrates strong independence enforcement, and GA provides balanced performance. The proposed framework achieves over 90% coverage with minimal node overhead, demonstrating scalability and robustness. This work establishes a foundation for hybrid and adaptive metaheuristic strategies in large-scale IoT security deployment.
Original Article
Control
Zahra Molavi-Nafchi; Abdorreza Rabiee; Maryam Shahriari
Abstract
In this paper, an adaptive observer-based cyber-attack detection and estimation problem is studied for smart grid under false data injection attack (FDIA). By invoking mathematical model of the smart grid, sliding mode observer (SMO) is proposed for estimating grid states and generating residual signal. ...
Read More
In this paper, an adaptive observer-based cyber-attack detection and estimation problem is studied for smart grid under false data injection attack (FDIA). By invoking mathematical model of the smart grid, sliding mode observer (SMO) is proposed for estimating grid states and generating residual signal. By monitoring grid status, evaluating residual signal and comparing it with the appropriate threshold level, attack detection is done and alarm signal is generated. Upon alarm generation, attack estimation algorithm is activated to estimate the FDIA occurred at the vulnerable buses. In the proposed detection scheme, only the frequency deviations of the generator buses are required; also, no off-line learning phase and no prior knowledge about attack signal are required. Moreover, the proposed scheme is able to detect FDIA and exactly estimates the severity of the detected attack. The Lyapunov stability theorem is used to guarantee stability of the proposed state observer and the proposed attack estimation algorithm, separately. Simulation results for the IEEE six-bus network under single-point and multi-point FDIAs show that upon attack occurrence, the defined evaluation function exceeds from a defined threshold level, which makes attack detection after some milliseconds. Also, the reported mean square error shows that the proposed attack estimation strategy can precisely estimate the attack shape and severity and identify the under-attack bus.
Research Articles
Optimization
Maryam Khalili Fard; Ali Hatami
Abstract
Fractional-order PID (FOPID) controllers offer enhanced flexibility and robustness compared to classical integer-order PID controllers, making them well suited for complex and nonlinear control systems. However, the optimal tuning of FOPID parameters remains a challenging task due to the increased number ...
Read More
Fractional-order PID (FOPID) controllers offer enhanced flexibility and robustness compared to classical integer-order PID controllers, making them well suited for complex and nonlinear control systems. However, the optimal tuning of FOPID parameters remains a challenging task due to the increased number of tuning parameters and the nonlinear nature of the resulting optimization problem. In this paper, a chaos-enhanced Grey Wolf Optimization (CGWO) algorithm is proposed for the optimal tuning of a FOPID controller applied to a DC motor system.Chaotic maps are integrated into the standard Grey Wolf Optimizer to improve population diversity, enhance global search capability, and mitigate premature convergence. Logistic and Tent chaotic maps are employed to dynamically regulate the control parameters of the optimizer throughout the optimization process. The proposed CGWO algorithm is implemented in a MATLAB/Simulink environment, and the controller performance is evaluated using the Integral of Time-Weighted Absolute Error (ITAE) as the optimization objective.Simulation results demonstrate that the proposed CGWO-FOPID controller outperforms both the conventional PID controller and the standard GWO-based FOPID controller in terms of transient response, convergence behavior, and control accuracy. Specifically, reduced overshoot, shorter settling time, and lower ITAE values are achieved. These results confirm that incorporating chaotic dynamics into metaheuristic optimization significantly enhances the tuning performance of fractional-order controllers and provides an effective and practical solution for industrial control applications.
