Computer Engineering Commons^™

Learning Target Class Eigen Subspace (Ltc-Es) Via Eigen Knowledge Grid, Sanjay Kumar Sonbhadra, Sonali Agarwal, P. Nagabhushan

Turkish Journal of Electrical Engineering and Computer Sciences

In one-class classification (OCC) tasks, only the target class (class-of-interest (CoI)) samples are well defined during training, whereas the other class samples are totally absent. In OCC algorithms, the high dimensional data adds computational overhead apart from its intrinsic property of curse of dimensionality. For target class learning, conventional dimensionality reduction (DR) techniques are not suitable due to negligence of the unique statistical properties of CoI samples. In this context, the present research proposes a novel target class guided DR technique to extract the eigen knowledge grid that contains the most promising eigenvectors of variance-covariance matrix of CoI samples. In …

An Adaptive Search Equation-Based Artificial Bee Colony Algorithm For Transportation Energy Demand Forecasting, Durmuş Özdemi̇r, Safa Dörterler

Turkish Journal of Electrical Engineering and Computer Sciences

This study aimed to develop a new adaptive artificial bee colony (A-ABC) algorithm that can adaptively select an appropriate search equation to more accurately estimate transport energy demand (TED). Also, A-ABC and canonical artificial bee colony (C-ABC) algorithms were compared in terms of efficiency and performance. The input parameters used in the proposed TED model were the official economic indicators of Turkey, including gross domestic product (GDP), population, and total vehicle kilometer per year (TKM). Three mathematical models, linear (A-ABCL), exponential (A-ABCE), and quadratic (A-ABCQ) were developed and tested. Also, economic variables were generated using the "curve fitting" technique to …

A New Effective Denoising Filter For High Density Impulse Noise Reduction, Iman Elawady, Caner Özcan

Turkish Journal of Electrical Engineering and Computer Sciences

Today, thanks to the rapid development of technology, the importance of digital images is increasing. However, sensor errors that may occur during the acquisition, interruptions in the transmission of images and errors in storage cause noise that degrades data quality. Salt and pepper noise, a common impulse noise, is one of the most well-known types of noise in digital images. This noise negatively affects the detailed analysis of the image. It is very important that pixels affected by noise are restored without loss of image fine details, especially at high level of noise density. Although many filtering algorithms have been …

Twitter Account Classification Using Account Metadata: Organizationvs. Individual, Yusuf Mucahi̇t Çeti̇nkaya, Mesut Gürlek, İsmai̇l Hakki Toroslu, Pinar Karagöz

Turkish Journal of Electrical Engineering and Computer Sciences

Organizations present their existence on social media to gain followers and reach out to the crowds. Social media-related tasks and applications, such as social media graph construction, sentiment analysis, and bot detection, are required to identify the entities' account types. Some applications focus on personal accounts, whereas others only need nonpersonal accounts. This paper addresses the account classification problem using only minimum amount of data, which is the metadata of the account's profile. The proposed approach classifies accounts either as organization or individual, in a language-independent manner, without collecting the accounts' tweet content. The model uses a long short term …

Blmdp: A New Bi-Level Markov Decision Process Approach To Joint Bidding Andtask-Scheduling In Cloud Spot Market, Mona Naghdehforoushha, Mehdi Dehghan Takht Fooladi, Mohammad Hossein Rezvani, Mohammad Mehdi Gilanian Sadeghi

Turkish Journal of Electrical Engineering and Computer Sciences

In the cloud computing market (CCM), computing services are traded between cloud providers and consumers in the form of the computing capacity of virtual machines (VMs). The Amazon spot market is one of the most well-known markets in which the surplus capacity of data centers is auctioned off in the form of VMs at relatively low prices. For each submitted task, the user can offer a price that is higher than the current price. However, uncertainty in the market environment confronts the user with challenges such as the variable price of VMs and the variable number of users. An appropriate …

Development Of A Control Algorithm And Conditioning Monitoring For Peak Load Balancing In Smart Grids With Battery Energy Storage System, Turhan Atici, Sezai̇ Taşkin, İbrahi̇m Şengör, Maci̇t Tozak, Osman Demi̇rci̇

Turkish Journal of Electrical Engineering and Computer Sciences

As the traditional electricity grid transitions to the smart grid (SG), some emerging issues such as increased renewable energy penetration in the power system that cause load unbalances require new control methods. Storage of energy seems to be the best option to struggle with such issues. In this manner, energy storage technologies ensure the operating flexibility of the distribution system operator in the power system in terms of both sustainability of energy and peak load balancing. In this study, a grid condition monitoring user-interface and control algorithm is developed for the peak load reduction and supply-demand balancing in a SG …

A Hybrid Acoustic-Rf Communication Framework For Networked Control Of Autonomous Underwater Vehicles: Design And Cosimulation, Saeed Nourizadeh Azar, Oytun Erdemi̇r, Mehrullah Soomro, Özgür Gürbüz Ünlüyurt, Ahmet Onat

Turkish Journal of Electrical Engineering and Computer Sciences

Underwater control applications, especially ones using autonomous underwater vehicles (AUVs) have become very popular for industrial and military underwater exploration missions. This has led to the requirement of establishing a high data rate communication link between base stations and AUVs, while underwater systems mostly rely on acoustic communications. However, limited data rate and considerable propagation delay are the major challenges for employing acoustic communication in missions requiring high control gains. In this paper, we propose a hybrid acoustic and RF communication framework for establishing a networked control system, in which, for long distance communication and control the acoustic link is …

Evaluating The Role Of Carbon Quantum Dots Covered Silica Nanofillers On The Partial Discharge Performance Of Transformer Insulation, Kasi Viswanathan Palanisamy, Chandrasekar Subramaniam, Balaji Sakthivel

Turkish Journal of Electrical Engineering and Computer Sciences

The article presents the experimental results on the role of carbon quantum dots (CQD) covered silica nanofillers on the partial discharge (PD) properties of transformer oil insulation. The improvement in PD performance of nanofiller blend oil is tested with increased voltage gradient and nanofiller concentration. PD of nanoblend oils for various concentrations of modified silica ranging from 0 to 0.1%wt was measured. PD activity of the test samples is simulated in the laboratory with needle, rod and plane electrode geometry combinations. The facets of PD signals such as PD magnitude, PD inception and time duration of PD extracted from phase-resolved …

A Novel Crimping Technique Approach For High Power White Good Plugs, Ömer Ci̇han Kivanç, Okan Özgönenel, Ömer Bostan, Şahi̇n Güzel, Mert Demi̇rsoy

Turkish Journal of Electrical Engineering and Computer Sciences

The crimping process is essential to human health and the durability of devices, especially in domestic appliances. Moreover, terminal crimping is critical to the safe transmission of electricity; incorrect crimping leads to problems including overheating of the plug, power loss, arc, and failure of the mechanical connection. In recent years, analysis has been performed by the finite element method (FEM) to prevent the incorrect design of crimping and to develop higher performance crimping techniques. A novel crimping technique for domestic appliances requiring high powered plugs is proposed in this study. After defining the crimp parameters and the materials that are …

A Novel Deep Reinforcement Learning Based Stock Price Prediction Using Knowledge Graph And Community Aware Sentiments, Anil Berk Altuner, Zeynep Hi̇lal Ki̇li̇mci̇

Turkish Journal of Electrical Engineering and Computer Sciences

Stock market prediction has been an important topic for investors, researchers, and analysts. Because it is affected by too many factors, stock market prediction is a difficult task to handle. In this study, we propose a novel method that is based on deep reinforcement learning methodologies for the prediction of stock prices using sentiments of community and knowledge graph. For this purpose, we firstly construct a social knowledge graph of users by analyzing relations between connections. After that, time series analysis of related stock and sentiment analysis is blended with deep reinforcement methodology. Turkish version of Bidirectional Encoder Representations from …

A New Speed Planning Method Based On Predictive Curvature Calculation For Autonomous Driving, Beki̇r Öztürk, Volkan Sezer

Turkish Journal of Electrical Engineering and Computer Sciences

As the number of vehicles in traffic is increasing day by day, the accident rates and driving effort are significantly raising. For this reason, ensuring safety and driving comfort is becoming more and more important. Driver assistance systems are the most common systems that are adopted for this purpose. With the development of technology, lane tracking support and adaptive cruise control systems are now being sold as standard equipment. More advanced research is being done for fully autonomous driving. One of the most critical parts of autonomous driving is speed profile planning. In this paper, a curvature-based predictive speed planner …

A Novel Fault Detection Approach Based On Multilinear Sparse Pca: Application Onthe Semiconductor Manufacturing Processes, Riadh Toumi, Yahia Kourd, Dimitri Lefebvre

Turkish Journal of Electrical Engineering and Computer Sciences

Batch processes are extremely important to researchers since they are widely used in many fields such as biochemistry, pharmacy, and semiconductors. The powerful batch detection method is critical to increase the performance of the overall equipment and to reduce the use of check wafers. Many techniques have been used in batch process monitoring. Among them, the multivariate statistical process control (MSPC) is very useful in batch process monitoring because of the large number of records data. Therefore, batch processes have certain characteristics, such as multimodal batch nonlinearity trajectories, which were challenged by these MSPCs. In this paper, a novel process …

Strategic Integration Of Battery Energy Storage And Photovoltaic At Low Voltage Level Considering Multiobjective Cost-Benefit, Samarjit Patnaik, Manas Nayak, Meera Viswavandya

Turkish Journal of Electrical Engineering and Computer Sciences

Renewable energy sources, such as solar photovoltaic (PV) systems and battery energy storage systems (BESS), help reduce greenhouse gas emissions while fulfilling the world?s growing energy demand. The inclusion of BESS reduces the peak hour demand, and control of charging and discharging of BESS can be economical for distributors facing time-based energy pricing. This paper discusses a novel multiobjective Horse herd optimisation algorithm (MOHHOA) approach, which is inspired by the social behaviour of horses in herds for PV and BESS optimal allocation in the radial distribution system. The proposed algorithm combines multiple benefits like benefits from economic gain per day, …

Two Person Interaction Recognition Based On A Dual-Coded Modified Metacognitive (Dcmmc) Extreme Learning Machine, Saman Nikzad, Afshin Ebrahimi

Turkish Journal of Electrical Engineering and Computer Sciences

Human action recognition has been an active research area for over three decades. However, state-of-the-art proposed algorithms are still far from developing error-free and fully-generalized systems to perform accurate interaction recognition. This work proposes a new method for two-person interaction recognition from videos, based on well-known cognitive theories. The main idea is to perform classification based on a theory of cognition known as dual coding theory. The theory states that human brain processes and represents two types of information to learn/classify data named analogue and symbolic codes, i.e. (verbal as analogue and visual as symbolic). To implement such a theory …

Offline Tuning Mechanism Of Joint Angular Controller For Lower-Limb Exoskeleton With Adaptive Biogeographical-Based Optimization, Mohammad Soleimani Amiri, Rizauddin Ramli

Turkish Journal of Electrical Engineering and Computer Sciences

Designing an accurate controller to overcome the nonlinearity of dynamic systems is a technical matter in control engineering, particularly for tuning the parameters of the controller precisely. In this paper, a tuning mechanism for a proportional-integral-derivative (PID) controller of lower limb exoskeleton (LLE) joints by adaptive biogeographical based-optimization (ABBO) is presented. The tuning of the controller is defined as an optimization problem and solved by ABBO, which is an iterative algorithm inspired by a blending crossover operator (BLX-?). The parameters of the migration change proportionally to the growth of iteration that conveys the error to rapid convergence by narrowing the …

Anomaly Detection In Rotating Machinery Using Autoencoders Based On Bidirectional Lstm And Gru Neural Networks, Krishna Patra, Rabi Narayan Sethi, Dhiren Kkumar Behera

Turkish Journal of Electrical Engineering and Computer Sciences

A time series anomaly is a form of anomalous subsequence that indicates future faults will occur. The development of novel techniques for detecting this type of anomaly is significant for real-time system monitoring. Several algorithms have been used to classify anomalies successfully. However, the time series anomaly detection algorithm was not studied well. We use a new bidirectional LSTM and GRU neural networks-based hybrid autoencoder to detect if a machine is operating normally in this research. An autoencoder is trained on a set of 12 features taken from healthy operating data gathered promptly after a planned maintenance period using vibration …

Self-Efficacy Development In Elementary-Aged Learners Through Dance As An Algorithmic Thinking Tool, Niva Shrestha

Honors Theses

The purpose of this research is to demonstrate the effectiveness of a transdisciplinary approach in teaching computational thinking through dance to elementary-aged learners, with primary attention to females. With limited literature available on how pre-adolescents begin to construct conceptions of computer science and other engineering domains, including potential career pathways, the incentive of this project was to leverage a day camp for about 20 rising 3rd - 5th-grade learners to assess their identity development in computer science. Modules that teach computational thinking through dance paired with Unruly splats (block-based programmable electronic gadgets) were implemented. By conducting pre-and post-surveys and a …

Comparing Actively Managed Mutual Fund Categories To Index Funds Using Linear Regression Forecasting And Portfolio Optimization, Luke Weiner

Industrial Engineering Undergraduate Honors Theses

The global investment industry offers a wide variety of investment products especially for individual investors. One such product, index funds, which are younger than actively managed mutual funds, have typically outperformed managed funds. Despite this phenomenon, investors have displayed a tendency to continue investing in actively managed funds. Although only a small percentage of actively managed funds outperform index funds, the costs of actively managed funds are significantly higher. Also, managed fund performances are most often determined by their fund category such as growth or real estate. I wanted to answer the following question for individual investors: can we …

Symbolic Ns-3 For Efficient Exhaustive Testing, Jianfei Shao

School of Computing: Dissertations, Theses, and Student Research

Exhaustive testing is an important type of simulation, where a user exhaustively simulates a protocol for all possible cases with respect to some uncertain factors, such as all possible packet delays or packet headers. It is useful for completely evaluating the protocol performance, finding the worst-case performance, and detecting possible design or implementation bugs of a protocol. It is, however, time consuming to use the brute force method with current NS-3, a widely used network simulator, for exhaustive testing. In this paper, we present our work on Sym-NS-3 for more efficient exhaustive testing, which leverages a powerful program analysis technique …

A Quantitative Comparison Of Algorithmic And Machine Learning Network Flow Throughput Prediction, Cayden Wagner

All Theses

Applications ranging from video meetings, live streaming, video games, autonomous vehicle operations, and algorithmic trading heavily rely on low latency communication to operate optimally. A solution to fully support this growing demand for low latency is called dual-queue active queue management (AQM). Dual-queue AQM's functionality is reduced without network traffic throughput prediction.

Perhaps due to the current popularity of machine learning, there is a trend to adopt machine learning models over traditional algorithmic throughput prediction approaches without empirical support. This study tested the effectiveness of machine learning as compared to time series forecasting algorithms in predicting per-flow network traffic throughput …