The text classification task has a wide range of application domains for distinct purposes, such as the classification of articles, social media posts, and sentiments. As a natural language processing application, machine learning and deep learning techniques are intensively utilized in solving such challenges. One common approach is employing the discriminative word features comprising Bag-of-Words and n-grams to conduct text classification experiments. The other powerful approach is exploiting neural network-based (specifically deep learning models) through either sentence, word, or character levels. In this study, we proposed a novel approach to classify documents with contextually enriched word embeddings powered by the neighbor words accessible through the trigram word series. In the experiments, a well-known web of science dataset is exploited to demonstrate the novelty of the models. Consequently, we built various models constructed with and without the proposed approach to monitor the models' performances. The experimental models showed that the proposed neighborhood-based word embedding enrichment has decent potential to use in further studies.

Electric motors are actively used in various industries. Acoustic noise is crucial in electric motors, and specific standards are depending on their application areas. Despite better performance, including acoustic noise, than brushed motors, brushless motors still generate acoustic noise due to their mechanical, electrical, and electronic components. This study investigated the impact of varying the switching frequency through the driver on the acoustic noise of an external rotor brushless DC motor. Tests were conducted on a surface-mounted magnet brushless motor with different switching frequencies, and detailed information about the sections of the control board governing the brushless motor was provided. The study includes measurements of motor speed, current, switching frequency, phase signal, and acoustic noise measured at two different locations in decibels. It was observed that the acoustic noise increased at specific switching frequencies. Furthermore, the variation in switching frequency also affected the heating due to losses in the switching elements. Increasing the switching frequency in the 12-28 kHz range reduced motor speed and the measured acoustic noise, while temperature increases were observed in various frequency ranges.

With the widespread integration of renewable energy systems (RES) into the electric grid, maintaining power quality within specified limits has become a major focus. The total harmonic distortion (THD) value of the currents transferred to the grid and directly affecting power quality in (RES) such as grid-connected photovoltaic (PV) and fuel cells (FC) should not exceed 5%. To achieve this objective, various filter topologies are used to interface between the inverters, which are components of PV and FC systems, and the grid.Although there are numerous filter topologies, L and LCL-type passive filters are the most commonly used topologies in grid-connected systems. This study analyzes the L and LCL-type filter topologies for the H-bridge neutral point clamped (HNPC) inverter. The study’s modeling was carried out in the MATLAB/Simulink®. By comparing the simulation results, L and LCL filters’ harmonic attenuation capabilities, maximum power transferred to the grid, and effects on grid current have all been investigated.

PV irrigation systems have begun to be used intensively today, as energy needs increase. In Partially Shaded Conditions (PSC), the efficiency of the PV system decreases significantly, and traditional Maximum Power Point Tracking (MPPT) algorithms become insufficient. On the other hand, traditional MPPT algorithms require sensors to measure the current and voltage of the PV system. In this study, a sensorless hybrid MPPT algorithm is proposed to reduce system costs and enable operation without the need for PV system data. A simulation study was conducted in the MATLAB/Simulink environment to examine the PV system. The proposed algorithm has been tested under four different PSC scenarios. PV system power, motor speed, and currents were examined under each condition. The high maximum power tracking performance of the proposed algorithm is demonstrated through simulation results. In the steady state, the lowest MPPT efficiency was 95.66%, whereas the highest MPPT efficiency was 99.9%. The MPPT algorithm completed in less than 2 seconds, with the first stage taking 1.3 seconds to reach most of the maximum PV system power. The second stage of the MPPT algorithm was used to achieve maximum power in a narrower area.

After a short historical background and mentioning common application areas and different clinical modalities of pulsed electric fields and the phenomena such as electro permeabilization and electroporation are introduced based on the concept of the basic mechanism. Subsequently, pulse generation (PG) and the adapted modular multilevel converter (MMC) topology that is to be covered are introduced including the operating principles and very basic analysis. Finally, performance of the proposed MMC with four half bridge submodules is tested on a prototype built in laboratory. A bidirectional voltage is produced with 80 V amplitude pulsating at 100 kHz operation frequency.

Çok seviyeli inverterler (ÇSİ), iyileştirilmiş çıkış dalga formu ve azaltılmış harmonik distorsiyon gibi avantajlarıyla dikkat çekmektedir. Ancak, ÇSİ sistemlerinde hala harmonik sorunu bulunmaktadır. Araştırmacılar, bu sorunu çözmek için seçici harmonik eliminasyonu-darbe genişliği modülasyonu (SHE-PWM) tekniğini kullanmıştır. Ancak, seçilen harmoniklerin ortadan kaldırılması için gereken SHE denklemleri karmaşık ve yoğun hesaplamalar gerektirir. Bu makalede, ÇSİ'lerde SHE denklemlerini çözmek için nispeten yeni bir Afrika Akbabaları Optimizasyonu (AVO) algoritması sunulmaktadır. AVO tabanlı SHE-PWM tekniği, 7, ve 11 seviyeli kaskad ÇSİ üzerinde test edilmiştir. Önerilen algoritma, 0.1 ile 1.0 modülasyon indeksi aralığında uygun çözümler bulmayı başarmıştır. 0.5 ile 1.0 arasındaki modülasyon indeksi aralığında seçilen harmoniklerin başarılı bir şekilde yok edilebildiği ve temel gerilimin %0.5'ten daha az hata ile kontrol edilebildiği gözlenmiştir.

Today, the use of small unmanned aerial vehicles (UAVs) has increased due to technological advances. There has been an interest in using multiple UAVs instead of a single UAV to accomplish a given mission. This is because there are many scenarios where the capabilities of a single UAV are inadequate due to certain constraints (battery capacity, time in the air). For this reason, swarm UAV studies have increased. A swarm UAV consists of a large number of UAVs cooperating to accomplish a specific mission. This study shows three-dimensional formation control of a swarm UAV system in an obstacle environment. A centralized control architecture is used in this process. All task assignments are made from a centralized system. The Artificial potential fields method creates the formation at the target point by avoiding obstacles. The study was carried out using the Robot Operating System (ROS). The methods were tested in Webots simulation environment. Crazyflie robots were used in the experiments. In the simulation environment, square, star and v formations were first tested in two dimensions. Then, as an example of three-dimensional formation, the cubic and pyramid formations were created and observed.

Due to increasing both safety and efficiency of the traffic, Vehicular Ad Hoc Network (VANET) is a promising technology of Intelligent Transport Systems (ITS). Unique characteristics of VANETs including high mobility and strict delay constraints, require new routing solutions specific to these networks to be proposed. As one of those solutions, topology-based routing approaches aim to find the shortest path by managing routing tables. In this paper, recent topology-based routing approaches for VANETs are investigated in detail. Proactive, reactive and hybrid solutions are compared with respect to their advantages, disadvantages, updating procedures and network sizes. This paper will shed light on future studies since it provides detailed information about the current status of the literature in topology-based routing approaches in VANETs.

While building smart living areas, the main purpose is to increase the quality and comfort of life without compromising health and safety. With the increase of new technologies used in growing and updated modern living areas, we witness that the needs are shaped and changed with this change. The primary purpose of each new technology is to offer more safety, health, comfort and convenience to the person, organization or structure that it addresses. The basic means of the connected vehicles has been to ensure the safety of the user in the best way, and people expect more than vehicle. With the widespread use of autonomous vehicles, that is without driver assistant, in the near future, vehicles are expected to connect and communicate with the ecosystem structured for the future and among themselves. Thanks to this communication that the vehicles have, the traffic flow on the route, the road condition, the location of the pedestrians around, possible malfunctions and failures that may occur and all such information will be shared by the vehicle with each other and the future ecosystem. On the other hand, while the vehicles are becoming more connected with any of useful technologies, thanks to the remote connection provided by their technology, manufacturers can easily develop their vehicles without recalling. In this study, information was given about the connected vehicles, the remote connection infrastructure and methods. The advantages provided by these systems are explained.

In the literature, interest in automatic control systems that do not require human intervention and perform at the desired level increases day by day. In this study, a Twin Delay Deep Deterministic Policy Gradient (TD3), a reinforcement learning algorithm, automatically controls a DC motor system. A reinforcement learning method is an approach that learns what should be done to reach the goal and observes the results that come out with the interaction of both itself and the environment. The proposed method aims to adjust the voltage value applied to the input of the DC motor in order to reach output with single input and single output structure to the desired speed.