Yazar "Kiselychnyk, Oleh" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim
    Yayın
    High-precision parameter identification for the lorenz system via starfish optimization algorithm
    (Institute of Electrical and Electronics Engineers Inc., 2025) Ekinci, Serdar; Türkeri, Cebrail; İzci, Davut; Kiselychnyk, Oleh; Bektaş Güneş, Burcu
    Parameter estimation plays a vital role in the accurate modeling and control of chaotic systems due to their inherent sensitivity to initial conditions and system parameters. This study introduces a novel approach for parameter identification in the Lorenz system using the starfish optimization algorithm (SFOA). The problem is formulated as a nonlinear optimization task, and the SFOA is employed to minimize the discrepancy between the observed and simulated trajectories. The proposed method is evaluated across multiple independent runs and benchmarked against several well-known optimization techniques. The results demonstrate that the SFOA achieves highly accurate and consistent parameter estimates, exhibiting superior robustness and convergence behavior compared to other methods. These findings suggest that SFOA is a promising candidate for addressing parameter estimation challenges in chaotic and nonlinear dynamical systems.
  • Küçük Resim
    Yayın
    Opposition-based starfish optimization algorithm for function optimization
    (Institute of Electrical and Electronics Engineers Inc., 2025) Ekinci, Serdar; Türkeri, Cebrail; İzci, Davut; Kiselychnyk, Oleh; Bektaş Güneş, Burcu
    This paper proposes a novel hybrid optimization algorithm, termed the opposition-based starfish optimization algorithm (OB-SFOA), designed for solving complex function optimization problems. By integrating the opposition-based learning (OBL) mechanism into the standard starfish optimization algorithm (SFOA), the proposed method aims to enhance population diversity, improve convergence speed, and reduce the risk of stagnation in local optima. The performance of OB-SFOA is evaluated using six widely used benchmark functions (Rosenbrock, Step, Schwefel, Penalized, Kowalik, and Shekel 5) covering diverse optimization characteristics including unimodality, multimodality, and fixed dimensionality. Comparative evaluations were conducted against standard SFOA, grey wolf optimizer (GWO), and RIME (an emerging physics-based algorithm). Results indicate that OB-SFOA consistently outperforms the competing algorithms in terms of average fitness, stability, and best-found solutions across all benchmark cases. These outcomes demonstrate the robustness and effectiveness of OB-SFOA as a promising candidate for general-purpose function optimization tasks.