Yazar "Akbulut, Osman" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim Yok
    Öğe
    Hybrid Intelligent Control System for Adaptive Microgrid Optimization: Integration of Rule-Based Control and Deep Learning Techniques
    (Mdpi, 2024) Akbulut, Osman; Cavus, Muhammed; Cengiz, Mehmet; Allahham, Adib; Giaouris, Damian; Forshaw, Matthew
    Microgrids (MGs) have evolved as critical components of modern energy distribution networks, providing increased dependability, efficiency, and sustainability. Effective control strategies are essential for optimizing MG operation and maintaining stability in the face of changing environmental and load conditions. Traditional rule-based control systems are extensively used due to their interpretability and simplicity. However, these strategies frequently lack the flexibility for complex and changing system dynamics. This paper provides a novel method called hybrid intelligent control for adaptive MG that integrates basic rule-based control and deep learning techniques, including gated recurrent units (GRUs), basic recurrent neural networks (RNNs), and long short-term memory (LSTM). The main target of this hybrid approach is to improve MG management performance by combining the strengths of basic rule-based systems and deep learning techniques. These deep learning techniques readily enhance and adapt control decisions based on historical data and domain-specific rules, leading to increasing system efficiency, stability, and resilience in adaptive MG. Our results show that the proposed method optimizes MG operation, especially under demanding conditions such as variable renewable energy supply and unanticipated load fluctuations. This study investigates special RNN architectures and hyperparameter optimization techniques with the aim of predicting power consumption and generation within the adaptive MG system. Our promising results show the highest-performing models indicating high accuracy and efficiency in power prediction. The finest-performing model accomplishes an R2 value close to 1, representing a strong correlation between predicted and actual power values. Specifically, the best model achieved an R2 value of 0.999809, an MSE of 0.000002, and an MAE of 0.000831.
  • Küçük Resim Yok
    Öğe
    Visualizing ordered bivariate data on node-link diagrams
    (Elsevier, 2023) Akbulut, Osman; Mclaughlin, Lucy; Xin, Tong; Forshaw, Matthew; Holliman, Nicolas S.
    Node-link visual representation is a widely used tool that allows decision-makers to see details about a network through the appropriate choice of visual metaphor. However, existing visualization methods are not always effective and efficient in representing bivariate graph-based data. This study proposes a novel node-link visual model - visual entropy (Vizent) graph - to effectively represent both primary and secondary values, such as uncertainty, on the edges simultaneously. We performed two user studies to demonstrate the efficiency and effectiveness of our approach in the context of static nodelink diagrams. In the first experiment, we evaluated the performance of the Vizent design to determine if it performed equally well or better than existing alternatives in terms of response time and accuracy. Three static visual encodings that use two visual cues were selected from the literature for comparison: Width-Lightness, Saturation-Transparency, and Numerical values. We compared the Vizent design to the selected visual encodings on various graphs ranging in complexity from 5 to 25 edges for three different tasks. The participants achieved higher accuracy of their responses using Vizent and Numerical values; however, both Width-Lightness and Saturation-Transparency did not show equal performance for all tasks. Our results suggest that increasing graph size has no impact on Vizent in terms of response time and accuracy. The performance of the Vizent graph was then compared to the Numerical values visualization. The Wilcoxon signed-rank test revealed that mean response time in seconds was significantly less when the Vizent graphs were presented, while no significant difference in accuracy was found. The results from the experiments are encouraging and we believe justify using the Vizent graph as a good alternative to traditional methods for representing bivariate data in the context of node-link diagrams.(c) 2023 The Authors. Published by Elsevier B.V. on behalf of Zhejiang University and Zhejiang University (http://creativecommons.org/licenses/by/4.0/).