Yazar "Er, Mehmet Bilal" seçeneğine göre listele
Listeleniyor 1 - 8 / 8
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe A Novel Approach for Motor Bearing Fault Detection Using EMD-Based Denoising and Detrended Fluctuation Analysis & LSTM Multimodal Hybrid Features with K-Means Clustering(Springer Heidelberg, 2026) Er, Mehmet Bilal; Koca, TarkanPurpose Engine bearings are among the most critical components in rotating machinery, and their failure can often lead to the entire system stopping. Therefore, accurate and reliable monitoring of bearings is crucial for predictive maintenance and fault prevention. However, noisy signals, variable operating conditions, and the simultaneous occurrence of multiple fault types make it difficult to achieve high accuracy in fault diagnosis. This study proposes a new hybrid approach to overcome these challenges. Methods The proposed method consists of five main stages. In the first stage, Empirical Mode Decomposition (EMD) based preprocessing is applied to remove noise and unwanted components from the vibration signals. In the second stage, feature extraction is performed from the processed signals using two different methods: Detrended Fluctuation Analysis (DFA) captures the fractal dynamics and statistical structure of the signals, while a Long Short-Term Memory (LSTM) network learns the complex time-dependent relationships of the signals. In the third stage, these obtained features are combined to obtain a more meaningful representation. In the fourth stage, the K-Means clustering method is used to group the features to increase the accuracy of the classifiers. In the final stage, three powerful classification algorithms, namely XGBoost, Random Forest, and Support Vector Machines (SVM), are used to classify bearing faults. The HUST engine bearing dataset used in this study enabled us to evaluate the performance of our method under realistic conditions because it contained data collected under different loads, speeds, and fault types. Results Experimental analyses showed that the proposed hybrid structure exhibited superior performance compared to single feature sets. In particular, the combined use of DFA and LSTM-based features increased the generalizability and diagnostic power of the system. Furthermore, classification accuracy was further improved thanks to the feature vectors supported by EMD preprocessing and K-Means clustering. The obtained results show that the best performance was achieved when DFA +LSTM features were used together with EMD and K-Means and with the XGBoost classifier. In this case, 98.46% accuracy, 97.20% sensitivity, 97.24% precision, and 97.27% F1-score values were achieved. Conclusion This study demonstrates that the proposed hybrid approach yields reliable results under different operating conditions and in data sets containing multiple fault types. The proposed method provides a robust and generalizable solution for predictive maintenance applications in industrial systems, making significant contributions to the safety and availability of rotating machinery.Öğe An advanced machine learning-based approach for accurate forecasting of solar photovoltaic energy production(Springer London Ltd, 2025) Koca, Tarkan; Er, Mehmet Bilal; Kisecok, BusraSolar energy has a strategic importance among renewable energy sources due to its high potential and environmental sustainability features. Increasing energy demand and environmental concerns necessitate accurate estimation of solar energy production. In this study, the daily energy output (kWh) of a solar photovoltaic (PV) system is estimated using real operational data obtained from the Solar Power Plant of Baykan Denim Company, located in Malatya province of Turkey, with an installed capacity of 7090.47 kWp. The dataset includes key parameters such as irradiance (Wh/m(2)), temperature (degrees C), performance ratio (%), and temperature-corrected performance ratio (%). Four regression algorithms; Linear Regression, LSTM (Long Short-Term Memory), Random Forest, and Extreme Gradient Boosting (XGBoost) were comparatively analyzed under different data splitting strategies (70-30, 80-20, and 10-fold cross-validation). The results reveal that XGBoost consistently outperforms the other algorithms, achieving the highest accuracy and lowest error values. Specifically, the XGBoost model with 10-fold cross-validation achieved MAE: 0.006628 kWh, MSE: 0.000126 (kWh)(2), RMSE: 0.011222 kWh, and R-2: 0.998277, indicating near-perfect prediction capability. These findings demonstrate the robustness of the proposed framework and highlight its potential to ensure continuity in solar energy production and support efficient energy management processes.Öğe Analysis and classification of the mobile molecular communication systems with deep learning(Springer Heidelberg, 2022) Isik, Ibrahim; Er, Mehmet Bilal; Isik, EsmeNano networks focused on communication between nano-sized devices (nanomachines) is a new communication concept which is known as molecular communication system (MCs) in literature. The researchers have generally used fixed transmitter and receiver for MCs models to analyze the fraction of received molecules and signal to interference rate etc. In this study, contrary to the literature, a mobile MC model has been used in a diffusion environment by using five bits. It is concluded that when the receiver and transmitter are mobile, distance between them changes and finally this affects the probability of the received molecules at the receiver. After the fraction of received molecules is obtained for different mobility values of Rx and Tx (Drx and Dtx), deep learning's bi-directional long short-term memory (Bi-LSTM) model is applied for the classification of Rx and Tx mobilities to find the best MC model with respect to fraction of received molecules. Finally it is obtained that when the mobilities of Rx and Tx increase, the fraction of received molecules also increases. Bi-LSTM model of Deep learning is used on a data set consisting of five classes. The suggested model's accuracy, precision, and sensitivity values are obtained as 98.05, 96.49, and 98.01 percent, respectively.Öğe Bacterial-Based Molecular Communication: Simulation of a Fixed and Receding Receiver Scenarios in Varied Viscosities and Environmental Conditions(Wiley-V C H Verlag Gmbh, 2025) Duman, Mustafa Ozan; Isik, Ibrahim; Isik, Esme; Er, Mehmet BilalThis study introduces a novel bacterial-based molecular communication (BBMC) model for nanoscale information exchange, harnessing the chemotactic behavior of Escherichia coli (E. coli). A comprehensive 3D simulation framework is developed to analyze the impact of key parameters diffusion coefficient (D), chemoattractant release rate (Q), receiver (RX) speed (u), and initial transmitter-receiver distance (d) on communication performance. Results indicate that lower D values enhance the formation of chemoattractant gradients, leading to improved signal clarity and efficiency. Conversely, higher RX speeds distort these gradients, increasing signal reach time and reducing success rates. Elevated Q values significantly broaden the sensing range and improve reliability, particularly over larger distances, though their effect is diminished at high RX speeds. Notably, success rates drop sharply as d approaches the theoretical sensing threshold, underscoring the critical need for parameter tuning. Experimental results validate these findings and reveal a threshold beyond which bacterial movement becomes random, limiting effective signal transmission. These insights contribute to optimizing BBMC systems for greater efficiency and reliability. Applications include targeted drug delivery, environmental biosensing, and synthetic biology, where precise bacterial signaling is essential. The study also demonstrates simulation as a scalable, cost-efficient alternative to experimental methods, addressing complexity and feasibility in real-world scenarios.Öğe Classification of Diffusion Constants of Transmitter and Receiver and Distance Between Them Using Mobile Molecular Communication via Diffusion Model(Springer Heidelberg, 2024) Er, Mehmet Bilal; Isik, Ibrahim; Kuran, Umut; Isik, EsmeMolecular communication (MC) holds promise for enabling communication in scenarios where traditional wireless methods may be impractical or ineffective, offering unique capabilities for a range of applications in both natural and engineered systems. In this research, a novel approach to MC is explored, diverging from the standard use of stationary transmitter and receiver models typically found in the field. The study introduces a dynamic MC model, where both the transmitter and receiver are mobile within a diffusion environment. This model operates using a 5-bit system. The key finding is that the mobility of these nanodevices alters their distance, which in turn impacts the likelihood of molecule reception at the receiver. The study employs deep learning techniques, specifically a combination of Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks, to categorize the mobility patterns of the receiver (Rx) and transmitter (Tx). By analyzing various mobility rates (Drx and Dtx) and distances between the Tx and Rx, the research successfully identifies the most efficient mobile MC model in terms of molecule reception rates. The use of Linear Support Vector Machine alongside the CNN and LSTM hybrid feature vector resulted in an 87.68% accuracy in predicting diffusion coefficients. Moreover, using a Cubic Support Vector with the same hybrid feature vector, the study achieved an 88.09% accuracy in estimating the distance between the transmitter and receiver. The study concludes that an increase in the mobilities of Rx and Tx correlates with a higher rate of molecule reception.Öğe LSTM Tabanlı Derin Ağlar Kullanılarak Diyabet Hastalığı Tahmini(2021) Er, Mehmet Bilal; Işık, İbrahimÖz: Diyabet, vücudun yeterli miktarda insülini üretmemesi veya iyi kullanamadığı durumda kan şekerinin normalin üstüne çıkması ile ortaya çıkan bir hastalıktır. Kan şekeri insanların ana enerji kaynağıdır ve bu enerji tüketilen yiyeceklerden gıdalardan gelir. Bu hastalık tedavi edilmez ise ölümcül olabilir. Ancak, erken tanı konulup tedaviye başlandığında tedavisi en olanaklı hastalıklardan biridir. Geleneksel diyabet teşhis süreci zorlu olduğundan, diyabetin klinik ve fiziksel verileri kullanılarak yapay sinir ağı, görüntü işleme ve derin öğrenme gibi sistemler kullanılarak hastalık teşhis edilebilmektedir. Bu araştırmada diyabet teşhisi için derin öğrenmeye dayalı bir model sunulmaktadır. Bu bağlamda Evrişimsel Sinir Ağı (ESA), Uzun Kısa Süreli Bellek (Long-short Term Memory Networks- LSTM) modelinin hibrit kullanımı sınıflandırma için tercih edilmiştir. Ayrıca ESA ve LSTM modelleri deneylerde ayrı ayrı kullanılmıştır. Önerilen modelin performansını değerlendirmek için literatürde yaygın olarak kullanılan Pima Indians Diabetes veri seti kullanılmıştır. En yüksek sınıflandırma başarısı %86,45 olarak ESA+LSTM modelinden elde edilmiştir.Öğe Parkinson's detection based on combined CNN and LSTM using enhanced speech signals with Variational mode decomposition(Elsevier Sci Ltd, 2021) Er, Mehmet Bilal; Isik, Esme; Isik, IbrahimParkinson's disease (PD) can cause many non-motor and motor symptoms such as speech and smell. One of the difficulties that Parkinson's patients can experience is a change in speech or speaking difficulties. Therefore, the right diagnosis in the early period is important in reducing the possible effects of speech disorders caused by the disease. Speech signal of Parkinson patients shows major differences compared to normal people. In this study, a new approach based on pre-trained deep networks and Long short-term memory (LSTM) by using melspectrograms obtained from denoised speech signals with Variational Mode Decomposition (VMD) for detecting PD from speech sounds is proposed. The proposed model consists of four steps. In the first step, the noise is removed by applying VMD to the signals. In the second step, mel-spectrograms are extracted from the enhanced sound signals with VMD. In the third step, pre-trained deep networks are preferred to extract deep features from the mel-spectrograms. For this purpose, ResNet-18, ResNet-50 and ResNet-101 models are used as pre-trained deep network architecture. In the last step, the classification process is occurred by giving these features as input to the LSTM model, which is designed to define sequential information from the extracted features. Experiments are performed with the PC-GITA dataset, which consists of two classes and is widely used in the literature. The results obtained from the proposed method are compared with the latest methods in the literature, it is seen that it has a better performance in terms of classification performance.Öğe Predictive Modeling of Bacteria-Based Nanonetwork Performance Using Simulation-Driven Machine Learning and Genetic Algorithm Optimization(Wiley-V C H Verlag Gmbh, 2026) Duman, Mustafa Ozan; Isik, Ibrahim; Er, Mehmet Bilal; Tagluk, Mehmet Emin; Isik, EsmeBacteria-based nanonetwork (BN) offers a biologically inspired solution for enabling information exchange between nanomachines (NMs) in environments where traditional communication methods are ineffective. This study presents a 2D simulation model of a BN system that captures the chemotactic behavior of a single Escherichia coli (E. coli) bacterium navigating from a transmitter (TX) toward a receiver (RX) under varying environmental conditions. Key parameters, which are chemoattractant release rate (Q), TX-RX distance (d), and bacterial lifespan (), are systematically varied to evaluate their impact on communication performance, measured in terms of reach time and success rate. To enable accurate performance prediction without the need for computationally expensive repeated simulations, an analytical model is constructed using various machine learning (ML) techniques, including Linear Regression (LR), Random Forest (RF), and Multi-Layer Perceptron (MLP). Hyperparameters of MLP are optimized using a Genetic Algorithm (GA), significantly enhancing predictive accuracy and training stability. The results demonstrate the effectiveness of integrating dynamic simulation with data-driven modeling and hyperparameter optimization to represent complex system behavior. This framework offers valuable design insights for BN system development and supports the creation of efficient, scalable nanonetworks.
