Yazar "Yilmaz, Nihat" seçeneğine göre listele
Listeleniyor 1 - 15 / 15
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Application of neural network prediction model to full-scale anaerobic sludge digestion(WILEY-BLACKWELL, 2011) Guclu, Dunyamin; Yilmaz, Nihat; Ozkan-Yucel, Umay G.BACKGROUND: Process modeling is a useful tool for description and prediction of the performance of anaerobic digestion systems under varying operation conditions. The objective of this study was to implement a model to simulate the dynamic behavior of a large-scale anaerobic sewage sludge digestion system. Artificial neural network (ANN) models using algorithms best suited to environmental problems (the Levenberg-Marquardt algorithm and the 'gradient descent with adaptive learning rate' back propagation algorithms) were used to model the anaerobic sludge digester of the Ankara Central Wastewater Treatment Plant (ACWTP) using dynamic data. RESULTS: Based on the relatively low mean square error (MSE), mean absolute error (MAE) and mean absolute percentage error (MAPE) and very high r values, ANN models predicted effluent volatile solid (VS) concentration and methane yield satisfactorily. Effluent VS and methane yields were predicted by the ANN using only conventional parameters such as pH, temperature, flow rate, volatile fatty acids, alkalinity, dry matter and organic matter. The best back propagation algorithm was the gradient descent with adaptive learning rate algorithm in both models. In the training of the neural network, four-fold cross-validation was used for validation of the model for better reliability. CONCLUSION: The proposed ANN models were shown to be capable of dynamically predicting the VS and CH(4) production rates for real system behavior. Only relatively simple monitoring parameters are needed to build the model for this complex anaerobic digestion process. (C) 2011 Society of Chemical IndustryÖğe Comparison of the SLAM algorithms: Hangar experiments(E D P SCIENCES, 2016) Korkmaz, Mehmet; Yilmaz, Nihat; Durdu, AkifThis study purposes to compare two known algorithms in an application scenario of simultaneous localization and mapping (SLAM) and to present issues related with them as well. Mostly used SLAM algorithms Extended Kalman Filter (EKF) and Unscented Kalman Filter (UKF) are compared with respect to the point of accuracy of the robot states, localization and mapping. Because of considering the most implementations in the previous studies, the simulation environments are chosen as big as possible to provide reliable results. In this study, two different hangar regions are tried to be simulated. According to the outcomes of the applications, UKF-based SLAM algorithm has superior performance over the EKF-based one, apart from elapsed time.Öğe Design and Actuator Selection of a Lower Extremity Exoskeleton(IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 2014) Onen, Umit; Botsali, Fatih M.; Kalyoncu, Mete; Tinkir, Mustafa; Yilmaz, Nihat; Sahin, YusufLower extremity exoskeletons are wearable robots that integrate human intelligence with the strength of legged robots. Recently, lower extremity exoskeletons have been specifically developed for transportation of disabled individuals. This paper summarizes the anthropomorphic design of a lower extremity exoskeleton named "walking supporting exoskeleton (WSE)." WSE has been developed to support some fundamental motions (walking, sitting, standing, etc.) of disabled individuals who lost leg muscular activities completely or partially. WSE has two degrees of freedom per leg which are powered by electrical actuators. This paper discusses critical design criteria considered in mechanical design and actuator selection of WSE.Öğe Embedded Computer System Based Mobile Robots(IEEE, 2014) Comlekciler, Ismail Taha; Yilmaz, NihatDesign and software development of embedded computer- based robotic systems have been the main aim of this paper. For this aim, the Lego Mindstorms NXT robot platform which is supported by Microsoft Robotics Studio (MSRS) software development environment and equipped with ARM7 microprocessor, are used. Software development studies are made on it. Embedded computer systems are analysed generally and programming of these systems which are used in robotic studies are detailed. Therefore, experiments are made on two different software development environments which are written as following. NXT program developed for Mindstorms, Microsoft Robotics Studio analysed and Microsoft Visual Studio. Microsoft Robotics Studio is software development environments which is object based and uses drag and drop methods. For easy and rapid software development, lots of robot producers give support with software package to Robotics Studio which have physical simulation environment.Öğe Feature Selection Method Based on Artificial Bee Colony Algorithm and Support Vector Machines for Medical Datasets Classification(HINDAWI LTD, 2013) Uzer, Mustafa Serter; Yilmaz, Nihat; Inan, OnurThis paper offers a hybrid approach that uses the artificial bee colony (ABC) algorithm for feature selection and support vector machines for classification. The purpose of this paper is to test the effect of elimination of the unimportant and obsolete features of the datasets on the success of the classification, using the SVM classifier. The developed approach conventionally used in liver diseases and diabetes diagnostics, which are commonly observed and reduce the quality of life, is developed. For the diagnosis of these diseases, hepatitis, liver disorders and diabetes datasets from the UCI database were used, and the proposed system reached a classification accuracies of 94.92%, 74.81%, and 79.29%, respectively. For these datasets, the classification accuracies were obtained by the help of the 10-fold cross-validation method. The results show that the performance of the method is highly successful compared to other results attained and seems very promising for pattern recognition applications.Öğe Force Feedback Control of Lower Extremity Exoskeleton Assisting of Load Carrying Human(TRANS TECH PUBLICATIONS LTD, 2014) Sahin, Yusuf; Botsali, Fatih Mehmet; Kalyoncu, Mete; Tinkir, Mustafa; Onen, Umit; Yilmaz, Nihat; Baykan, Omer KaanLower extremity exoskeletons are wearable robot manipulators that integrate human intelligence with the strength of legged robots. Recently, lower extremity exoskeletons have been specifically developed for rehabilitation, military, industrial applications and rescuing, heavy-weight lifting and civil defense applications. This paper presents controller design of a lower-extremity exoskeleton for a load carrying human to provide force feedback control against to external load carried by user during walking, sitting, and standing motions. Proposed exoskeleton system has two legs which are powered and controlled by two servo-hydraulic actuators. Proportional and Integral (PI) controller is designed for force control of system. Six flexible force sensors are placed in exoskeleton shoe and two load cells are mounted between the end of the piston rod and lower leg joint. Force feedback control is realized by comparing ground reaction force and applied force of hydraulic cylinder. This paper discusses control simulations and experimental tests of lower extremity exoskeleton system.Öğe Formation Morphing of Multi-Robots Using Graph Theory: Fugitive Chasing(INT ASSOC ENGINEERS-IAENG, 2015) Erdogan, Kemal; Korkmaz, Mehmet; Durdu, Akif; Yilmaz, Nihat; Topal, SebahattinIn this study, it is considered the case of chasing escapers using graph theory method with multi robots, which are connected to each other. In this simulation study the fugitives are trying to escape from a campus region which has five possible gates for entrance and exit. Because of not determining the possible escape ways of fugitives, it is hard to obtain useful results from police or security chase in a quick time. In this approach, some security robots are waiting in stand-by position to get a new command for positioning in a specified area. Happening of any undesired case, robots are informed by an operator which gate is the alarmed gate. At this point, multi-robots are positioned by using graph theory if the communication range is appropriate to connect each other. This imaginary scenario is supposed to take place in the campus of Selcuk University. There are five gates in the campus that are used for vehicle and pedestrian entrance and exit. For any unusual events similar to this simulation like shooting or robbery, security robots would be alarmed. After alarm case, the backup team would be sent to the gates to start the aerial chasing of escapers.Öğe A hybrid breast cancer detection system via neural network and feature selection based on SBS, SFS and PCA(SPRINGER, 2013) Uzer, Mustafa Serter; Inan, Onur; Yilmaz, NihatTwo hybrid feature selection methods (SFSP and SBSP) which are composed by combining the sequential forward selection and the sequential backward selection together with the principal component analysis developed by utilizing quadratic discriminant analysis classification algorithmic criteria so as to utilize in the diagnosis of breast cancer fast and effectively are presented in this study. The tenfold cross-validation method has been applied in the algorithm, which is utilized as criteria during the selection of the features. The dimension of the feature space for input has been decreased from 9 to 4 thanks to the selection of these two hybrid features. The Artificial Neural Networks have been used as classifier. The cross-validation method has been preferred also in the phase of this classification as in the case of the selection of the feature in order to increase the reliability of the result. The Wisconsin Breast Cancer Database obtained from the UCI has been utilized so as to determine the correctness of the system suggested. The values of the average correctness of the classification obtained by utilizing a tenfold cross-validation of the two hybrid systems developed earlier are found, respectively, as follows: for SFSP + NN, 97.57 % and for SBSP + NN, 98.57 %. SBSP + NN system has been observed that, among the studies carried out by implementing the cross-validation method for the breast cancer, the result appears to be very promising. The acquired results have revealed that this hybrid system applied by means of reducing dimension is an utilizable system in order to diagnose the diseases faster and more successfully.Öğe Intention Recognition Using Leap Motion Controller and Artificial Neural Networks(IEEE, 2016) Erdogan, Kemal; Durdu, Akif; Yilmaz, NihatIntention recognition is an important topic in the field of Human Robot Interaction. If the robot is wanted to make counter movements just in time according to human's actions, a robotic system must recognize the intention of the human necessarily. In this paper, a method for a robotics system to estimate the human's intention is presented. In our method, the information is provided from the sensor called as leap motion controller device. The decision about the tendency of human intention is made by Artificial Neural Network. A scenario has been designed that a human subject tries to pile the boxes on each other. The main point for this robotic system and the scenario is to recognize the intention as which box would be held by the subject.Öğe Mechanical Design of Lower Extremity Exoskeleton Assisting Walking of Load Carrying Human(TRANS TECH PUBLICATIONS LTD, 2014) Sahin, Yusuf; Botsali, Fatih Mehmet; Kalyoncu, Mete; Tinkir, Mustafa; Onen, Umit; Yilmaz, Nihat; Cakan, AbdullahExoskeletons are used in rehabilitation, military, industrial applications and rescuing, heavy-weight lifting and civil defense applications as well. This paper presents to design of a lower-extremity exoskeleton assisting walking of a load carrying human. Proposed exoskeleton system is designed to be appropriate mechanism with human lower extremity and it operates synchronously with the human realizes. The aim of exoskeleton actuator system is to provide forces against to external load carried by user during walking, sitting, and standing motions. Thus, it supports human walking and significant portion of external load carrying by the user. Also it makes possible to user spend less energy, less stress and fatigue. Proposed work involves the following design steps: kinematic synthesis of the exoskeleton, mechanical and electro-hydraulic system design.Öğe A New Data Preparation Method Based on Clustering Algorithms for Diagnosis Systems of Heart and Diabetes Diseases(SPRINGER, 2014) Yilmaz, Nihat; Inan, Onur; Uzer, Mustafa SerterThe most important factors that prevent pattern recognition from functioning rapidly and effectively are the noisy and inconsistent data in databases. This article presents a new data preparation method based on clustering algorithms for diagnosis of heart and diabetes diseases. In this method, a new modified K-means Algorithm is used for clustering based data preparation system for the elimination of noisy and inconsistent data and Support Vector Machines is used for classification. This newly developed approach was tested in the diagnosis of heart diseases and diabetes, which are prevalent within society and figure among the leading causes of death. The data sets used in the diagnosis of these diseases are the Statlog (Heart), the SPECT images and the Pima Indians Diabetes data sets obtained from the UCI database. The proposed system achieved 97.87 %, 98.18 %, 96.71 % classification success rates from these data sets. Classification accuracies for these data sets were obtained through using 10-fold cross-validation method. According to the results, the proposed method of performance is highly successful compared to other results attained, and seems very promising for pattern recognition applications.Öğe ON-LINE A PREDICTIVE MODEL OF CUTTING FORCE IN TURNING WITH 3 AXIS ACCELERATION TRANSDUCER USING NEURAL NETWORK(AMER SOC MECHANICAL ENGINEERS, 2009) Asilturk, Ilhan; Yilmaz, NihatIn this study cutting forces prediction was modeled using back propagation (BP) neural network algorithm. Experimental turning dataset is used in this study to train and evaluate the model. The Input dataset includes speed, feed rate, depth of cut, vibration levels along the three axes on tool holder (ax,ay,az). The Output dataset includes feed force, vertical force, and radial force. Marginally acceptable results were given by early experiments of this study and when data was examined, high non-linearity can be seen from the prepared graphic.. In the previous work, a fine development of reliability of predicting the cutting forces can be observed by the help of results. To compare the estimated results of cutting force from this method with the cutting force signal can be measured directly by dynamometer; it is found that the difference between measured and estimated cutting forces is less than 0.2% in all case.Öğe Real-Time Line Tracking Based on Web Robot Vision(WILEY-BLACKWELL, 2011) Yilmaz, Nihat; Sagiroglu, SerefThis study presents a real-time line tracking application based on web robot vision. The application is implemented with the help of Web-SUN robot platform specially designed for research and development. Real-time images were processed with the support libraries. The processes such as monitoring, tele-control, parameter adjustment, and remote-reprogramming were implemented on the designed platform through standard Internet browser without any additional software. A projection-based line tracking algorithm and conventional PID control algorithm are used in study. Line tracking experiments are performed on a test track drawn by red tape on floor which includes curves and corners in laboratory environment. In this study, our real-time line tracking experiments was completed smoothly in 10 tours on the test track with a maximum deviation of 12 cm. (C) 2009 Wiley Periodicals, Inc. Comput Appl Eng Educ 19: 806-813, 2011; View this article online at wileyonlinelibrary.com/journal/cae; DOI 10.1002/cae.20367Öğe Web robot learning powered by Bluetooth communication system(IEEE COMPUTER SOC, 2006) Sagiroglu, Seref; Yilmaz, Nihat; Wani, M. ArifThis paper presents a web robot web-robot learning powered by Bluetooth communication system. The web-robot system is used as the virtual robot laboratory integrating a number of disciplines in engineering. This virtual laboratory is a valuable teaching tool for engineering education used at any time and from any location through Internet. The mobile robot was controlled with robot server named as control center The server can be connected to mobile robot via Bluetooth adapter The mobile robot system focuses on vision sensing. Real time image processing techniques are realized by the web robot system. This system can also realize monitoring, tele-controlling, parameter adjusting and reprogramming through Internet exclusively with a standard Web browser without the need of an v additional software.Öğe Web-based maze robot learning using fuzzy motion control system(IEEE COMPUTER SOC, 2007) Yilmaz, Nihat; Sagiroglu, SerefIn this study, a web based maze robot system has been designed and implemented for solving different maze algorithms with the help of machine learning approaches. The robot system has a map-based heuristic maze solving algorithm. The algorithm used for solving the maze is based on map creation and produces a control signal for robot direction. Robot motions were controlled by a fuzzy motion control system running on a chip. The control algorithm can be easily changed with the help of an algorithm via web interface controlled by the control center. The control center program powered by AM TLAB functions and special libraries (image and control) in DELPHI manage all robotic activities. These activities are: command interpreter, image capturing, processing and serving, machine learning techniques, web serving, database management, communication with robot, and compiling microcontroller programs. The results have shown that the proposed designed and implemented system provides amazing new features to the applicants doing their real-time programming exercises on web.
