Meltem Yavuz Çelikdemir

Meltem Yavuz Çelikdemir


Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2025 in IET Renewable Power Generation
Verified
Journal Article
Open access
MÇMeltem Yavuz Çelikdemir
Meltem Yavuz Çelikdemir
The reliable and efficient operation of polymer electrolyte membrane fuel cells (PEMFCs) necessitates the implementation of preventive strategies and maintenance protocols to minimize the likelihood of failures. To address this, the study identifies effective diagnostic techniques for detecting faults in PEMFCs. A data‐driven approach leveraging machine learning methods is proposed to enhance the detection of flooding faults under varying operational conditions. This approach enables the automatic extraction of fault‐related features directly from raw data. Experimental data obtained from an 80 W PEMFC, widely used in the literature for comparability, was utilized in the study. Various machine learning classification algorithms were applied, and their performance metrics were analysed. Among these, the random subspace k‐nearest neighbor algorithm demonstrated superior accuracy and the shortest training time, leading to the development of a novel model. The study evaluated 22 variables associated with PEMFCs, performed fault diagnosis, and assessed fault severity. Furthermore, a risk analysis was conducted using the proposed model, enabling the prediction of both the risk level and the probability of fault occurrence as percentages. Key performance metrics, including accuracy, sensitivity, precision, and specificity, were calculated as 99.97%, 99.98%, 99.90%, and 99.98%, respectively, during model validation. During testing, these metrics were recorded as 99.45%, 100%, 98.42%, and 99.16%, respectively.

30 May 2025 in International Journal of Energy and Smart Grid
Verified
Journal Article
Open access
PDF file
MÇMeltem Yavuz Çelikdemir
Meltem Yavuz Çelikdemir
This study aims to provide a more realistic and sustainable perspective for estimating the investment costs of hydroelectric power plants within the framework of the Clean Development Mechanism (CDM). For the first time in the literature, the CDM parameter has been directly integrated into an empirical equation model, and the model parameters have been optimized using the Particle Swarm Optimization (PSO) algorithm. The proposed model was tested with technical and economic data from various hydroelectric power plants in India, achieving high accuracy in cost estimation by considering both installed capacity and the level of CDM implementation. The results demonstrate that the proposed model operates with lower error rates compared to conventional methods and enables a quantitative analysis of the impact of sustainability-oriented policy instruments on energy costs. In this study, 30 hydroelectric power plants in India were analyzed, and highly accurate cost estimations were obtained using the proposed equation model in conjunction with the optimization algorithm, resulting in a total error of 4.49 and an average error of 0.14. In this respect, the study offers a significant methodological contribution to the integration of sustainability criteria into investment planning and policy development processes in the energy sector.

24 January 2025 in Journal of Polytechnic
Verified
Journal Article
Open access
PDF file
MÇMeltem Yavuz Çelikdemir
Meltem Yavuz Çelikdemir
AAAyhan Akbal
Ayhan Akbal
Cerebral aneurysms are an important disease that threatens human life. Rupture of these aneurysms causes hemorrhages in the cerebral arteries. Computed Tomography Angiography (CTA) is widely used in the clinical diagnosis of cerebral aneurysms. Interpretation errors by radiologists in examining these Computed Tomography (CT) images are vital for patients. Based on this importance, deep learning-based studies aim to help keep these errors to a minimum. For this purpose, CTA images were used to detect cerebral aneurysms in this study. For CTA image analysis, deep learning methodology was preferred through Convolutional Neural Network (CNN). The validation accuracy of the training obtained as a result of deep learningg has a high rate of validation with 99.54% accuracy, 100% sensitivity, 98.89% specificity and 100% precision. As a training dataset, it yielded 127 true positives and 1 false positive for patient images with aneurysm, 89 true negatives and 0 false negative for images of patients with non-aneurysms. In this trained network, results were obtained with a high accuracy of 86.6% on 75 CTA images for external test. Regional dimension analysis was also performed for an image with an aneurysm detected in the test process.

30 April 2025 in Traitement du Signal
Verified
Journal Article
Open access
MÇMeltem Yavuz Çelikdemir
Meltem Yavuz Çelikdemir
AAAyhan Akbal
Ayhan Akbal

16 December 2025 in Muş Alparslan Üniversitesi Fen Bilimleri Dergisi
Verified
Journal Article
Closed access
MÇMeltem Yavuz Çelikdemir
Meltem Yavuz Çelikdemir
AAAyhan Akbal
Ayhan Akbal
Cerebral aneurysms are a major, life threatening cerebrovascular disease, and accurate interpretation of Computed Tomography Angiography (CTA) is critical for early diagnosis and treatment. This study evaluates the effectiveness of deep learning in reducing radiologist related interpretation errors by applying 15 different Convolutional Neural Networks (CNNs) to 1,211 CTA images. Prior to classification, images underwent various preprocessing and filtering operations, and comparative performance metrics were obtained. The best result, representing the highest accuracy reported to date of 99.72%, was achieved with a smoothing filtered image dataset using the VGG19 architecture. In the VGG19 test set, model outputs consisted of 272 true negatives (tn), 1 false negative (fn), 0 false positives (fp), and 90 true positives (tp). These findings demonstrate that appropriate image preprocessing and filtering significantly enhance CNN based aneurysm detection performance and play a crucial role in improving classification accuracy.