Yazar "Smerat, Aseel" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • Küçük Resim Yok
    Öğe
    Blue-eared Hedgehog Optimization (BEHO): A Nature-inspired Metaheuristic for Robust and Efficient Global Optimization
    (Intelligent Network and Systems Society, 2025) Dinler, Özlem Batur; Bektemyssova, Gulnara; Ahmed, Mahmood Anees; Ibraheem, Ibraheem Kasim; Smerat, Aseel; Montazeri, Zeinab; Eguchi, Kei
    A novel metaheuristic algorithm named the Blue-Eared Hedgehog Optimization (BEHO), inspired by the unique foraging and defensive behaviour of the blue-eared hedgehog is introduced in this study. Unlike conventional optimization methods, BEHO simulates the species’ natural strategies-nocturnal cautious exploration, gradual environmental mapping, and protective retreat-into computational operators that effectively balance exploration and exploitation. The algorithm initializes a diverse population of candidate solutions, simulates hedgehog-inspired gradual movements for exploration, and employs defensive-inspired refinement for exploitation, ensuring robust convergence and preservation of high-quality solutions. BEHO’s performance has been rigorously evaluated on 23 standard benchmark functions, including unimodal, high-dimensional multimodal, and fixed-dimensional multimodal problems, and compared with nine state-of-the-art metaheuristics, including MOA, WaOA, AOA, GWO, LSA, SWO, TLBO, BaOA, and WSO. Experimental results demonstrate that BEHO consistently achieves superior accuracy, stability, and convergence speed across all function categories. Its hedgehog-inspired mechanisms allow the algorithm to escape local optima, maintain population diversity, and achieve precise global solutions in complex and high-dimensional landscapes. The findings highlight BEHO as a highly effective and versatile optimization tool, providing a biologically grounded and computationally efficient framework for solving diverse complex problems. © This article is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. License details: https://creativecommons.org/licenses/by-sa/4.0/
  • Küçük Resim Yok
    Öğe
    Carpenter Optimization Algorithm: A Human-inspired Metaheuristic for Robust and Efficient Constrained Optimization
    (Intelligent Network and Systems Society, 2025) Dinler, Özlem Batur; Bektemyssova, Gulnara; Şahin, Canan Batur; Montazeri, Zeinab; Dehghani, Mohammad; Smerat, Aseel; Eguchi, Kei
    This paper introduces the Carpenter Optimization Algorithm (COA), a novel human-inspired metaheuristic designed to efficiently solve complex, high-dimensional, and constrained optimization problems. COA draws direct inspiration from the systematic behaviors of skilled carpenters, who initially perform broad cuts to explore raw materials and subsequently execute precise refinements to achieve high-quality outcomes. These behaviors are mathematically mapped into exploration and exploitation phases, where stochastic global modifications mimic broad exploratory actions, and targeted incremental adjustments refine promising solutions. Unlike traditional metaheuristics, COA achieves a robust balance between exploration and exploitation while requiring minimal control parameters, enhancing both adaptability and computational efficiency. The algorithm was rigorously evaluated on 22 constrained benchmark functions from the CEC 2011 suite and compared against nine well-established metaheuristics. The results demonstrate that COA consistently outperforms all competitors in terms of solution quality, convergence speed, stability, and robustness, achieving the best mean, median, and best values across all test problems. Statistical analyses, including standard deviation, rank-based evaluation, and pairwise Wilcoxon tests, confirm the significance and reproducibility of these results, while boxplot visualizations highlight controlled variability and narrow interquartile ranges, even for large-scale and multimodal problems. The findings suggest that COA’s behaviorally grounded design provides a practical and explainable framework for real-world optimization tasks. Future research directions include extending COA to multi-objective, dynamic, and large-scale industrial problems, integrating hybrid strategies or adaptive mechanisms, and conducting theoretical analyses of convergence and parameter sensitivity. Overall, COA represents a promising addition to the metaheuristic optimization landscape, offering both conceptual clarity and high practical performance. This article is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. License details: https://creativecommons.org/licenses/by-sa/4.0/
  • Küçük Resim Yok
    Öğe
    Detection of cyber attacks in electric vehicle charging systems using a remaining useful life generative adversarial network
    (Nature Portfolio, 2025) Tanyildiz, Hayriye; Sahin, Canan Batur; Dinler, Ozlem Batur; Migdady, Hazem; Saleem, Kashif; Smerat, Aseel; Abualigah, Laith
    Cybersecurity attacks targeting electric vehicle supply equipment (EVSE) and the broader electric vehicle (EV) ecosystem have become an escalating concern with the increasing adoption of EVs and the growing connectivity of the infrastructure supporting them. The present research aims to contribute to continuing cybersecurity studies on electric vehicle charging stations. In line with this objective, this study proposes the remaining useful life (RUL) approach to demonstrate the potential impact of estimating the remaining time of a cyber attack on EVSE and what revolutionary changes it can bring to cyber security strategies using a generative adversarial network (GAN). By taking a proactive stance, the manuscript will increase security and reduce the economic and reputational losses associated with cyber incidents. Accurate RUL estimates present valuable information about the status of the EVSE infrastructure. Thus, informed decisions on maintenance and crew scheduling are taken. To test the technique's effectiveness, we assess this approach on attack scenarios, including network and host attacks on the EV charger (Electric Vehicle Supply Equipment-EVSE) in idle and charging states. Furthermore, we assess the prediction results of different deep learning models, such as gated recurrent units (GRUs), long short-term memory (LSTM), recurrent neural networks (RNNs), convolution neural networks (CNNs), multi-layer perceptron (MLP), and dense layer integrated with generative adversarial networks (GANs), using mean absolute error (MAE), root mean square error (RMSE), mean squared error (MSE), and R-squared (R2). Afterward, we compare the error measurements with models, such as hybrid GAN-LSTM, GAN-GRU, GAN-RNN, GAN-CNN, GAN-MLP, and GAN-Dense Layer. The GAN-GRU model exhibits the highest accuracy with the lowest MAE (0.0281). On the contrary, the GAN-CNN model displays the best overall performance concerning error consistency and variance explained. According to the results, integrating GAN into these architectures improves predictive accuracy and the model's ability to identify potential attacks in advance and decreases error rates.