Yıldız Technical University Research Information System
Not a member yet
91324 research outputs found
Sort by
A memristive synaptic circuit and optimization algorithm for synaptic control
In order for the backpropagation training method, which is widely used for machine learning inference layer, to be directly applied to memristor crossbar arrays, either the weight change must be linear, or since the memristance change is not constant over time, the current memristance value must be kept in memory or changes must be controlled with an algorithm suitable for the used memristance function. To overcome the memory and energy drawbacks of this non-linearity, in this study, the parameters of a memristive circuit that can implement positive and negative weights were determined by the optimization method, using two charge-controlled mathematial memristor equations and a flux-controlled memristor emulator previously defined in the literature. In this way, the simplest linear control of weight change is achieved. Using the artificial bee colony algorithm, the passive element values of a circuit that can perform weight control up to 0.02 sensitivity and the duration of the applied control signal were determined. According to the experimental study, it was seen that weight control was achieved with a mean square error of 2.33×10−4. Also the tracking rate of software-based test accuracy is 98.186%. With the proposed optimization method and cost function, linear control can be achieved by determining the parameters needed for online training with any memristor element
SOSYAL DUYGUSAL ÖĞRENME MESLEKİ GELİŞİM EĞİTİMİNİN TASARLANMASI UYGULANMASI VE DEĞERLENDİRİLMESİ
Simple Determination of Cystine in Artificial Urine and Drug Matrices Using Digital Image Colorimetry
Smartphone-based digital image colorimetry was developed for thedetermination of cystine in pharmaceutical formulations and artificialurine samples. A lab-made colorimetric box was designed for captur-ing the digital images of sample solutions using a smartphone. Thedeveloped method employs a simple colorimetric method callednitroprusside test which specifically targets free -SH groups. Cystinewas reduced to cysteine before the colorimetric test was applied.The color intensity is inversely related to the concentration of theanalyte. The best linearity was obtained with the G color channel.Digital image processing was performed using the RGB ColorDetector mobile application to obtain analytical data. The developedmethod exhibited good linearity in the range of 29.7–00.3 mg/kg.The limits of detection and quantification were 9.7 mg/kg and32.3 mg/kg, respectively. The percent recoveries were 104.2 ± 7.8%and 108.7 ± 4.6%, which proved the accuracy and applicability of thedeveloped method</p
Environmentally Sustainable and Socially Resilient Urban Development and Management
This study investigates the impact of built environment features and vegetation on microclimatic conditions in Yıldızkent, a neighborhood of Erzurum, a cold-climate city in Turkey. Our research entails first creating one baseline and 20 different urban design scenarios with various built environment parameters and green spaces. Second, microclimatic conditions resulting from urban design scenarios were simulated for summer and winter using the ENVI-met program to test the impact of each identified parameter.The results show that building density and cluster type significantly affect the microclimatic conditions in summer, while an increase in green space is much more effective in regulating the temperature in winter. For some cities such as Erzurum, the microclimate comfort level is reached for them by cutting down the BCR by 50%, applying the linear-courtyard cluster type, as well as incorporating 50% of green areas. The solution is to decrease the values of BCR in 1.5 times compared to the primary scenario with maintaining the cluster type linear-courtyard layout for increased comfort of microclimate. For regions with cold climates, a linear-courtyard cluster type building organization is more advantageous than a molecular layout. Increasing the green area can greatly enhance microclimate comfort, especially during the winter season.</p
A Fuzzy Risk Assessment Approach Based on Z-Numbers for Enhancing Safety and Human-Robot Collaboration in Automotive Sector
Industry 5.0 (I5.0) technologies introduce new workplace safety challenges, particularly in human–robot collaboration environments. While robots handle nonergonomic, repetitive, and hazardous tasks, existing risk assessment methods often fail to address the uncertainties in dynamic human–robot interactions. To bridge this gap, this study proposes a novel risk evaluation framework integrating fuzzy set theory with Z-numbers. The methodology integrates Delphi method for expert consensus on risk factors, decision-making trial and evaluation laboratory for causal relationships, analytic network process for importance considering interdependencies, and VIseKriterijumska Optimizacija I Kompromisno Resenje for ranking risks to prioritize mitigation actions. The methodology uniquely addresses hesitancy of experts’ judgments and data imprecision through a systematic approach validated through a case study in one of Turkey's leading commercial vehicle manufacturers on a bus production line utilizing gantry-type industrial robots. A sensitivity analysis using VIKOR parameters further validates robustness. The Z-number framework overcomes traditional risk scores by differentiating scenarios yielding similar scores but distinct profiles, distinguishing low-probability/high-severity hazards from high-probability/low-severity ones, leading to nuanced prioritization
İstanbul İlinin İlçe Bazlı Deprem Risk Haritasının Oluşturulması
Depremler, dünyanın dinamik yapısından kaynaklanan veinsan hayatı ile altyapıyı olumsuz etkileyen en yıkıcı doğal afetlerdenbiridir. Depremlerin sıklığı, büyüklüğü ve sonuçları, coğrafi konum ve yer altıyapıları gibi çeşitli faktörlere bağlıdır. Türkiye, özellikle İstanbul, KuzeyAnadolu Fay Hattı üzerinde yer aldığı için yüksek deprem riski altındadır. Bunedenle, şehrin etkili hazırlık stratejileri geliştirmesi büyük önemtaşımaktadır. İstanbul, Türkiye'nin en büyük şehri olup, nüfus ve ekonomikfaaliyetler bakımından önemli bir konumda yer alırken, tarihsel ve coğrafideğerleriyle de öne çıkmaktadır. Tarih boyunca İstanbul, yıkıcı depremleryaşamıştır ve 1999 İzmit Depremi, şehrin yakınlarında meydana gelerekgelecekteki büyük bir felaketin potansiyelini gözler önüne sermiştir. Bu durum,deprem risklerinin değerlendirilmesi ve etkilerin azaltılmasına yönelikstratejilerin geliştirilmesinin gerekliliğini vurgulamaktadır. Bu çalışma,İstanbul'daki deprem risklerini analiz etmek amacıyla Analitik Hiyerarşi Süreci(AHP) ve Ideale En Yakın Çözüm Tekniği (TOPSIS) yöntemlerini birleştirmektedir.AHP, kriterlerin göreli önemini değerlendirmek için kullanılırken, TOPSIS, İstanbul’unilçelerini deprem riski açısından sıralamak için kullanılmaktadır. Buçalışmanın amacı, İstanbul'un farklı bölgelerindeki deprem risklerinideğerlendirmek ve ilçeleri risk açısından sıralamaktır. AHP ve TOPSISyöntemlerini kullanarak, objektif bir analiz sunulmakta ve şehir plancıları,politika yapıcılar ve karar vericiler için değerli bilgiler sağlanmaktadır.Çalışmanın bulguları, İstanbul'un deprem hazırlığını artırabilecek etkili riskazaltma stratejilerinin geliştirilmesine katkı sağlamayı hedeflemektedir.</p
Fredholmness of the Dirichlet Problem for 2mth-Order Elliptic Equations in Grand Sobolev Spaces
In this paper, on a bounded domain (Formula presented.) with a sufficiently smooth boundary (Formula presented.), it is considered a uniformly elliptic equation of (Formula presented.) th order, the coefficients of which are continuous in the principal part. Grand Lebesgue space (Formula presented.), is studied. This space is nonseparable, and it is defined a separable subspace (Formula presented.) of (Formula presented.), in which infinitely differentiable functions are dense. Furthermore, the grand Sobolev space (Formula presented.) of (Formula presented.) th-order differentiable in the Sobolev sense functions is introduced. This space is generated by the subspace (Formula presented.). For the given equation, a Schauder-type estimate up to the boundary is proved. Using this estimate, we establish an a priori estimate and then the Fredholmness of the (Formula presented.) th-order elliptic equation under consideration in (Formula presented.). By a solution, we mean a strong solution