Graduate School of Engineering and Science

Permanent URI for this communityhttps://hdl.handle.net/10679/8952

Browse

Now showing 1 - 11 of 11

Open Access
Critical review of the parameters affecting the effectiveness of moisture absorption treatments used for natural composites
(MDPI AG,, 2019) Al-Maharma, Ahmad Yousef Mohammad; Al-Huniti, N.; Al-Maharma, Ahmad Yousef Mohammad
Natural composites can be fabricated through reinforcing either synthetic or bio-based polymers with hydrophilic natural fibers. Ultimate moisture absorption resistance at the fiber–matrix interface can be achieved when hydrophilic natural fibers are used to reinforce biopolymers due to the high degree of compatibility between them. However, the cost of biopolymers is several times higher than that of their synthetic counterparts, which hinders their dissemination in various industries. In order to produce economically feasible natural composites, synthetic resins are frequently reinforced with hydrophilic fibers, which increases the incompatibility issues such as the creation of voids and delamination at fiber–matrix interfaces. Therefore, applying chemical and/or physical treatments to eliminate the aforementioned drawbacks is of primary importance. However, it is demonstrated through this review study that these treatments do not guarantee a sufficient improvement of the moisture absorption properties of natural composites, and the moisture treatments should be applied under the consideration of the following parameters: (i) type of hosting matrix; (ii) type of natural fiber; (iii) loading of natural fiber; (iv) the hybridization of natural fibers with mineral/synthetic counterparts; (v) implantation of nanofillers. Complete discussion about each of these parameters is developed through this study.
Open Access
Data fusion analysis and synthesis framework for improving disaster situation awareness
(MDPI, 2023-09) Aksit, M.; Say, Hanne; Eren, Mehmet Arda; de Camargo, V. V.; Say, Hanne; Eren, Mehmet Arda
To carry out required aid operations efficiently and effectively after an occurrence of a disaster such as an earthquake, emergency control centers must determine the effect of disasters precisely and and in a timely manner. Different kinds of data-gathering techniques can be used to collect data from disaster areas, such as sensors, cameras, and unmanned aerial vehicles (UAVs). Furthermore, data-fusion techniques can be adopted to combine the data gathered from different sources to enhance the situation awareness. Recent research and development activities on advanced air mobility (AAM) and related unmanned aerial systems (UASs) provide new opportunities. Unfortunately, designing these systems for disaster situation analysis is a challenging task due to the topological complexity of urban areas, and multiplicity and variability of the available data sources. Although there are a considerable number of research publications on data fusion, almost none of them deal with estimating the optimal set of heterogeneous data sources that provide the best effectiveness and efficiency value in determining the effect of disasters. Moreover, existing publications are generally problem- and system-specific. This article proposes a model-based novel analysis and synthesis framework to determine the optimal data fusion set among possibly many alternatives, before expensive implementation and installation activities are carried out.
Open Access
Deep learning-based expressive speech synthesis: a systematic review of approaches, challenges, and resources
(Springer, 2024-02-12) Barakat, Huda Mohammed Mohammed; Turk, O.; Demiroğlu, Cenk; Electrical & Electronics Engineering; DEMİROĞLU, Cenk; Barakat, Huda Mohammed Mohammed
Speech synthesis has made significant strides thanks to the transition from machine learning to deep learning models. Contemporary text-to-speech (TTS) models possess the capability to generate speech of exceptionally high quality, closely mimicking human speech. Nevertheless, given the wide array of applications now employing TTS models, mere high-quality speech generation is no longer sufficient. Present-day TTS models must also excel at producing expressive speech that can convey various speaking styles and emotions, akin to human speech. Consequently, researchers have concentrated their efforts on developing more efficient models for expressive speech synthesis in recent years. This paper presents a systematic review of the literature on expressive speech synthesis models published within the last 5 years, with a particular emphasis on approaches based on deep learning. We offer a comprehensive classification scheme for these models and provide concise descriptions of models falling into each category. Additionally, we summarize the principal challenges encountered in this research domain and outline the strategies employed to tackle these challenges as documented in the literature. In the Section 8, we pinpoint some research gaps in this field that necessitate further exploration. Our objective with this work is to give an all-encompassing overview of this hot research area to offer guidance to interested researchers and future endeavors in this field.
Open Access
Deep transformer-based asset price and direction prediction
(IEEE, 2024) Gezici, Abdul Haluk Batur; Sefer, Emre; Computer Science; SEFER, Emre
The field of algorithmic trading, driven by deep learning methodologies, has garnered substantial attention in recent times. Within this domain, transformers, convolutional neural networks, and patch embedding-based techniques have emerged as popular choices within the computer vision community. Here, inspired by the latest cutting-edge computer vision methodologies and the existing work showing the capability of image-like conversion for time-series datasets, we apply more advanced transformer-based and patch-based approaches for predicting asset prices and directional price movements. The employed transformer models include Vision Transformer (ViT), Data Efficient Image Transformers (DeiT), and Swin. We use ConvMixer for a patch embedding-based convolutional neural network architecture without a transformer. Our tested transformer-based and patch-based methodologies aim to predict asset prices and directional movements using historical price data by leveraging the inherent image-like properties within the historical time-series dataset. Before the implementation of attention-based architectures, the historical time series price dataset is transformed into two-dimensional images. This transformation is facilitated through the incorporation of various common technical financial indicators, each contributing to the data for a fixed number of consecutive days. Consequently, a diverse set of two-dimensional images is constructed, reflecting various dimensions of the dataset. Subsequently, the original images depicting market valleys and peaks are annotated with labels such as Hold, Buy, or Sell. According to the experiments, trained attention-based models consistently outperform the baseline convolutional architectures, particularly when applied to a subset of frequently traded Exchange-Traded Funds (ETFs). This better performance of attention-based architectures, especially ViT, is evident in terms of both accuracy and other financial evaluation metrics, particularly during extended testing and holding periods. These findings underscore the potential of transformer-based approaches to enhance predictive capabilities in asset price and directional forecasting. Our code and processed datasets are available at https://github.com/seferlab/price_transformer.
Open Access
Depression screening from voice samples of patients affected by parkinson’s disease
(S. Karger AG, 2019-05-01) Özkanca, Yasin Serdar; Öztürk, M. G.; Ekmekci, Merve Nur; Atkins, D. C.; Demiroğlu, Cenk; Ghomi, R. H.; Electrical & Electronics Engineering; DEMİROĞLU, Cenk; Özkanca, Yasin Serdar; Ekmekci, Merve Nur
Depression is a common mental health problem leading to significant disability worldwide. It is not only common but also commonly co-occurs with other mental and neurological illnesses. Parkinson's disease (PD) gives rise to symptoms directly impairing a person's ability to function. Early diagnosis and detection of depression can aid in treatment, but diagnosis typically requires an interview with a health provider or a structured diagnostic questionnaire. Thus, unobtrusive measures to monitor depression symptoms in daily life could have great utility in screening depression for clinical treatment. Vocal biomarkers of depression are a potentially effective method of assessing depression symptoms in daily life, which is the focus of the current research. We have a database of 921 unique PD patients and their self-assessment of whether they felt depressed or not. Voice recordings from these patients were used to extract paralinguistic features, which served as inputs to machine learning and deep learning techniques to predict depression. The results are presented here, and the limitations are discussed given the nature of the recordings which lack language content. Our models achieved accuracies as high as 0.77 in classifying depressed and nondepressed subjects accurately using their voice features and PD severity. We found depression and severity of PD had a correlation coefficient of 0.3936, providing a valuable feature when predicting depression from voice. Our results indicate a clear correlation between feeling depressed and PD severity. Voice may be an effective digital biomarker to screen for depression among PD patients.
Open Access
The effect of interface gradient distribution on unrealistic flow in 3D droplet simulations
(Europe, Institute for Liquid Atomization and Spray Systems, ILASS, 2021-08-31) Yılmaz, Anıl; Kayansalçik, Gökhan; Ertunç, Özgür; Mechanical Engineering; ERTUNÇ, Özgür; Yılmaz, Anıl; Kayansalçik, Gökhan
The purpose of this study is to investigate the origin of the parasitic current to provide accurate prediction of droplet surface interactions in Volume of Fluid (VOF) framework. The deformation of the droplet due to parasitic current has been the most important problem in 3D simulations. Parasitic current is influenced by curvature and surface normal estimation in the Continuum Surface Force (CSF) model. It has been shown that the number of neighboring cells of the central cell influences the gradient calculations regarding the generation of parasitic current. It has been observed that the polyhedral cell structure delivers a smoother interface gradient distribution than the cartesian cell structure. To examine the dynamics in different physical conditions, we compared simulations with base experiments to understand whether those models work. We then simulated droplet cases on stationary and moving wall conditions, and simulation results were consistent with experimental results.
Open Access
A multi-depot vehicle routing problem with time windows for daily planned maintenance and repair service planning
(Pamukkale Üniversitesi, 2023) Toru, Elif; Yılmaz, G.; Toru, Elif
A compressor manufacturer producing in Kocaeli/Dilovasi region makes vehicle routing and employee planning daily to fulfill the maintenance and repair requests of the Marmara region and its surroundings the next day. The service types and times are agreed upon with the customer before service planning. The vehicles and their respective operators for a given planning day are known, with the service personnel's starting and ending points being the residences. All the planned services must be satisfied in the time windows customers give. We approach the issue as a multi-depot vehicle routing problem with time windows (MDVRPTW) and construct a mixed-integer linear programming framework. The solution is deemed adequate in resolving the company's service planning predicament. To tackle large instances, we formulate a clustering algorithm that yields a proficient solution in a concise duration.
Open Access
A predictive multistage postdisaster damage assessment framework for drone routing
(Wiley, 2024-01) Adsanver, Birce; Göktürk, Elvin Çoban; Koyuncu, Burcu Balçık; Industrial Engineering; GÖKTÜRK, Elvin Çoban; Adsanver, Birce
This study focuses on postdisaster damage assessment operations supported by a set of drones. We propose a multistage framework, consisting of two phases applied iteratively to rapidly gather damage information within an assessment period. In the initial phase, the problem involves determining areas to be scanned by each drone and the optimal sequence for visiting these selected areas. We have adapted an electric vehicle routing formulation and devised a variable neighborhood descent heuristic for this phase. In the second phase, information collected from the scanned areas is employed to predict the damage status of the unscanned areas. We have introduced a novel, fast, and easily implementable imputation policy for this purpose. To evaluate the performance of our approach in real-life disasters, we develop a case study for the expected 7.5 magnitude earthquake in Istanbul, Turkey. Our numerical study demonstrates a significant improvement in response time and priority-based metrics.
Open Access
Towards interactive explanation-based nutrition virtual coaching systems
(Springer, 2024-01) Buzcu, Berk; Tessa, M.; Tchappi, I.; Najjar, A.; Hulstijn, J.; Calvaresi, D.; Aydoğan, Reyhan; Computer Science; AYDOĞAN, Reyhan; Buzcu, Berk
The awareness about healthy lifestyles is increasing, opening to personalized intelligent health coaching applications. A demand for more than mere suggestions and mechanistic interactions has driven attention to nutrition virtual coaching systems (NVC) as a bridge between human–machine interaction and recommender, informative, persuasive, and argumentation systems. NVC can rely on data-driven opaque mechanisms. Therefore, it is crucial to enable NVC to explain their doing (i.e., engaging the user in discussions (via arguments) about dietary solutions/alternatives). By doing so, transparency, user acceptance, and engagement are expected to be boosted. This study focuses on NVC agents generating personalized food recommendations based on user-specific factors such as allergies, eating habits, lifestyles, and ingredient preferences. In particular, we propose a user-agent negotiation process entailing run-time feedback mechanisms to react to both recommendations and related explanations. Lastly, the study presents the findings obtained by the experiments conducted with multi-background participants to evaluate the acceptability and effectiveness of the proposed system. The results indicate that most participants value the opportunity to provide feedback and receive explanations for recommendations. Additionally, the users are fond of receiving information tailored to their needs. Furthermore, our interactive recommendation system performed better than the corresponding traditional recommendation system in terms of effectiveness regarding the number of agreements and rounds.
Open Access
Two-part bio-based self-healing repair agent for cement-based mortar
(International Center for Numerical Methods in Engineering, 2020) Tezer, Mustafa Mert; Bundur, Zeynep Başaran; Civil Engineering; BUNDUR, Zeynep Başaran; Tezer, Mustafa Mert
Factors affecting durability of concrete structures are generally associated with each other. Due to its brittle nature, concrete can crack under stress and these cracks are one of the main reasons for a decrease in service life in concrete structures. Therefore, it is crucial to detect and recover microcracks, then to repair them as they were developed to wider cracks. Recent research in the field of concrete materials suggested that it might be possible to develop a smart cement-based material that is capable of remediate cracks by triggering biogenic calcium carbonate (CaCO3) precipitaton. This paper summarizes a study undertaken to investigate the self-healing efficiency of Sporosarcina pasteurii (S. pasteurii) cells immobilized on both diatomaceous earth and pumice, to remediate flexural cracks on mortar in early ages (28 days after mixing). To obtain a two-phase bio additive, half of the minerals were saturated with a nutrient medium consisting of urea, corn-steep liqueur(CSL) and calcium acetate and the cells with immobilized to the other half without nutrients. Screening of the healing process was done with ultrasonic pulse velocity (UPV) testing and stereomicroscopy. With this approach, the cracks on mortar surface were sealed and the water absorption capacity of the so-called self-healed mortar decreased compared to its counterpart cracked mortar samples.
Open Access
Validation and comparison of 2D and 3D numerical simulations of flow in simplex nozzles
(Europe, Institute for Liquid Atomization and Spray Systems, ILASS, 2021-08-31) Bal, M.; Kayansalçik, Gökhan; Ertunç, Özgür; Böke, Y. E.; Mechanical Engineering; ERTUNÇ, Özgür; Kayansalçik, Gökhan
Numerical simulations of pressure swirl atomizers are computationally expensive due to transient and multiphase flow behavior. In this study, 2D and 3D VOF simulations are performed for a geomerty which has high swirl chamber length-to-diameter ratio of 1.33. discharge coefficient (CD) and spray angle values are compared to the experimental data. Moreover, a benchmark study is conducted between 2D and 3D methods in terms of accuracy, computational cost and flow variables such as orifice exit axial and tangential velocity. The simulations are performed using a hybrid RANS-LES approach, IDDES model. It is observed that 2D simulation has lower accuracy in the validation parameters such as discharge coefficient and spray angle as compared to the 3D simulation. The main reason for 2D simulation inaccuracy might be the tangential port inlet effects and wrong estimation of the loss of swirl inside the swirl chamber. On the other hand, 2D simulations have approximately 1000 times lower computational cost than 3D simulations.

Browse

Browsing by Rights "info:eu-repo/semantics/openAccess"