Theses supervised by Prof. Dr. Hakan Yavuz
12 theses · Çukurova University
Geni̇şleşti̇ri̇lmi̇ş kontrol aralikli manyeti̇k retarder
Vehicles often use mechanical braking systems. In these brake systems, the kinetic energy of the vehicle is converted into heat and are released into the medium by brake system components. Although some serious difficulties are encountered in mechanical brake system, they are very much compact and effective in most cases. In particular, increasing temperature reduces brake forces posing a serious problem for the vehicle control. In addition, the mechanical brake system's complexity, the contamination of the brake fluid with water may lead to serious technical problems. The targeted magnetic retarder system is expected to resolve some of these issues and is intended to provide an alternative braking system. In this type of magnetic retarder based brake systems, the kinetic energy of the vehicle is not converted into heat by friction. On the other hand, in the standard magnetic retarder system, there are 4 brake force levels in total. The targeted system has been designed to increase this brake force levels to 1024. Instead of conventional four positions, the new retarder force control arm is therefore could be positioned anywhere between 0 and 1024. The retarder lever arm position is used in generation of the magnetic retarders electronically controlled brake force. In this manner, a highly sensitive desired level of braking force can be achieved.
Design and development of a light weight composite chassis for electric vehicle applications
Today, the vehicle technology is being transformed due to environmental impact of fossil fuel and their uses. Especially studies appear to be directed towards development of lightweight and durable chassis system. In this way, along with achieving fuel economy in existing systems, the need for lightweight chassis for electrically powered vehicles are met. In recent years, stress analysis studies using the finite element method has made significant contributions to the design engineering. With the help of a computer software, stress and deformation analysis of a system can be achieved with very small margin of errors at very much realistic values. As a part of the study, some sample carbon fiber-epoxy composite material was manufactured and tested. Considering the test results obtained, a car chassis for two passengers is designed using SolidWorks software. The developed composite and aluminum based chassis have been analyzed on the ANSYS software and the development of the chassis has been completed successfully.
Command generation techniques for elimination of residual vibrations for robotic applications
One method used to reduce or eliminate residual vibrations is to modify the input signal by using previously determined system parameters. In order to eliminate the residual vibration completely, these system parameters must be known very accurately. In real systems, achieving such accuracy may not always be possible. To address this problem and to provide a solution, a new residual vibration elimination method is introduced in this thesis. It has been proven to be useful especially in cases of uncertain system parameters that are estimated or predicted. It is shown that the technique is capable of handling high levels of uncertainty and is able to successfully eliminate or reduce residual vibrations in flexible robotic systems. In this approach, the desired position of the system is primarily divided into two or more equal parts for each of which a sectional input command is generated. In generation of each of the input signal, cycloid-plus-ramped and versine-plus-ramp functions are used. This reference input is composed of three key functions. Each part is travelled by each of the three functions within the same travel time. Provided that the specified move time and the total distance are unchanged, vibration can be eliminated by adjusting excursion distance of each function. The thesis presents theoretical and experimental results of the techniques applied to flexible robotic systems; a comparative study of robustness performance is also provided. The presented simulation and experimental results show that the residual vibrations are considerably decreased with a high degree of robustness in the presence of estimation related uncertainty regarding system parameters.
Parametric design of a semi-trailer chassis
This study examines the application of parametric design of a semi-trailer chassis with the purpose of adding flexibility to the design process required by fast evolving transportation technology. Parametric design allows the design of parts that are similar but have different geometric measurements to be made in a short time and to pass the designs onto production quickly. Catia V5 design software was used for software development. In Catia V5, parametric design is done by using VBA because user defined forms can be utilized in code writing. The design environment is a structure in which we can control the whole semi-trailer chassis within the framework of the designer specified rules. In the system, the basic parameters that make up the chassis are entered by the user and the chassis is dimensioned and then the drawing, modelling, static and dynamic analysis of the chassis can be performed automatically in the software environment. This work allowed a standard design that meets certain criteria for design, allowing the revision to be faster and more functional compared to conventional design techniques.
Static analysis of two seater extended range electric vehicle
Electrical driven vehicles are becoming an attractive alternative to combustion engine cars. Electric vehicles will play quite an important role in reducing emissions in the future. One of the most effective ways for reducing fuel consumption and emissions can be achieved by reducing weight of the vehicle. The vehicle chassis is one of the most important components of vehicle structure because of its versatile role on the dynamic behavior of the vehicle. There is a great potential for weight reduction of the chassis. This study presents the design and development of a two-seat prototype electric vehicle chassis. The main objectives are designing a battery module that can be easily attached and detached to the battery module compartment of the chassis, designing a new chassis model and ultimately evaluating the chassis for deformation based on static analysis of the entire structure. Vehicle designs and analyzes were done with Catia V5 R20 software. AL6061, ST-37 and carbon fiber materials are preferred for chassis.
Use of artificial intelligence in design of input shaper for motion control of mechanical systems
The cranes are usually controlled by a variety of actions including handling, lifting, carrying, then lowering and gripping the load. An intelligent control system could provide the operator with important advantages when performing such operations. In this study, it is shown how to perform an effective transport operation under variable conditions with hybrid control methods on a simple pendulum. Zero vibration input shaping method is implemented to a simple pendulum experiment set with different heights and variable weights. Single- Hidden-Layer Neural Network (NN), General Regression NN (GRNN), and Radial Basis Function Network (RBFN) are used in a serial configuration as an intelligent control method. Performance comparisons of these smart control methods are also made for different weight and height values other than the training data. Also, in order to operate such systems safely and economically, an effective automation system is needed too. In this study the results of automation system and mechanical damping methods have been presented. The mechanical damping method is necessary not only used for positioning a load with minimum oscillation but also for balancing it against wind forces. The experimental performances of these methods for different conditions have been compared with each other. In addition, current state-of-the-art industrial applications of automation and mechanical damping methods are also examined.
Investigation of the effect of the plasma actuators in aerodynamic flow control applications
This study aims at investigating the applications AI and feedback control assisted plasma actuators based active flow control of selected airfoil types for improvement of aerodynamic performances. In aerial systems, the plasma actuators have been shown to achieve promising results for active flow control. In the first part of the study, aerodynamic performance of the NACA 2415 and NACA 0015 airfoils are investigated with the implementation of plasma actuator for flow control. In second part of the study, a feedback control method is implemented to achieve active control of the plasma actuators. Also, artificial neural network technique is utilized to estimate the experimental data related to lift and drag coefficients needed for the aerodynamic performance analysis. The experimental results of the first part indicate that the plasma actuators located at suction surface of the airfoils provide decrease in drag force, increase in lift force, shifting of stall angle, performance enhancement of the airfoils and reduction of the wake width of the airfoils. The experimental results of the second part show that developed Artificial Neural Network model is successful in estimation of lift and drag coefficients. In addition, the study on feedback control algorithm to control the flow around the airfoils with the plasma actuators as an active flow control with a feedback controller has proven to have considerable benefits.
Parametric design of electrical and mechanical structure for a distribution transformer
Typically mechanical and electrical designs of distributed transformers are performed by using a 3D software packages. The parametric design modelling is used to create different variations of the same concept design. For this purpose, some special software settings and related features are used. These software features are 3D design program configurations, related i-logic features and visual basic encoding with Excel and some analysis programs. In this study, a hermetic and oil expansion tank type distributed transformers with a power range between from 25 kVA to 4000 kVA is being parametrically designed and developed. Firstly, parameterization process is performed on the main concept model to make variants of the main model. In order to parameterize a mechanical model, it is necessary to clarify the characteristic parameters of the model and how these parameters should be varied to achieve the variants. Usually, the characteristic parameters are more than one and some of them have a specific range. Therefore, the point to be considered here is that they should be used and examined in such a way that they do not impair the flexibility of the model. Hence, it is necessary to define parameters correctly and add design rules in order to ensure to keep the final product keep within the standards. After drawing the model, all dimensions must be converted into mathematical formulas by using macros and VBA coding for automatically updating for each design case. After entering initial requirements and data analysis stage, firstly dimensions of electrical parts are to be calculated depending upon some rules and formulas.Then, the transformer is drawn in generally to give numerical dimensions on software. Afterwards, the parameters are assigned to each dimension by using Excel data sets. All parameters and rules can be formed easily before drawings just once all necessary details are defined in Excel data. Finally, different variants of the design can be generated automatically just by defining the relevant deign related critical parameters are defined.
Control techniques for oscillating wave energy converter
This study aims at achieving increase in the power capture performance of heaving type oscillating wave energy converters with the proposed control techniques. It has been shown that the power capture performance of a heaving wave energy converter depends on meeting the resonance condition with the incident wave frequency. In regular sea wave conditions, the task of setting the control parameters is a well-known task where the wave frequency is clearly defined. However, in irregular sea conditions, the wave frequency needed for control settings is not clearly defined and there does not exist a widely accepted procedure for identification of wave frequency. In the first part of the study, an investigation on how the power capture performance of a heaving wave energy converter varies with the varying the ratio of power take off unit damping to radiation damping is performed. This study has proven to have considerable benefits. In second part of the study, a control algorithm including Fuzzy Logic controller which controls the added mass is proposed. The results of the study show that the power capture performance of the heaving wave energy conversion system is increased. Also, as the last part of this study, an approach is proposed that combine renewable energy production system with the reverse osmosis desalination plant. The power captured by the wave energy conversion system is utilized to produce fresh water from the reverse osmosis-based desalination system model. The simulation results indicates that the operation of the combined system is a feasible solution to the fresh water issue and it is also much environmentally friendly.
Investigation of process parameters in direct material deposition based open-source additive manufacturing systems
In this study, the Additive Manufacturing methods are investigated. They are considered to be revolutionary for the manufacturing industry. In this context, amongst many other research topics, it is observed that Fused Filament Fabrication and Metal Droplet Jetting systems are considered to be popular ones. The common feature of these methods is that they are based on the direct material deposition technique, which provides significant advantages in terms of manufacturing parts with complex geometry. However, both of the methods are fairly new and there are still some technical issues that are currently a research topic. To achieve the objectives of the presented study, firstly, a Fused Filament Fabrication device based on open-source concept is developed. The process parameters that affects the part strength are determined and optimized using the open-source device. During the optimizations, along with others, the issue of material consumption is also considered. The objective of this study is to achive controlled process parameters and steps to increase the part strength using the same amount of material. In the second stage; Owing to the open-source concept system developed, the same device is extensively modified to make it suitable for the Metal Droplet Jetting system related research. The production process is performed by controlled deposition of metal droplets. In order to give droplet form to the molten metal, a unique Metal Droplet Jetting mechanism including a mechanical actuation system is designed and manufactured. To achieve a deposition model with the desired morphology, the droplet ejection and deposition stages are first subdivided into groups, then each is examined in details. In this way, the droplets are not only produced homogeneously, but the need for pneumatic gas support, which makes it difficult to control and increases the processing cost, is eliminated. As a result of experimental studies supported by theoretical calculations, a low-cost and practical metal material deposition method that can be used effectively, especially in the electronics industry, is presented.
Design and control of a novel Eddy current dynamometer
The recent studies on Artificial Intelligence accompanied by enhanced computing capabilities supports increasing attention into traditional control methods coupled with AI learning methods in an attempt to bringing adaptiveness and fast responding features. The Model Predictive Control (MPC) and Proportional-Integrative-Differential (PID) control techniques are widely used. They are safe and reliable control method mainly based on constraints. On the other hand, the Eddy Current dynamometers are highly nonlinear braking systems whose performance parameters are related to many processes related variables. This study is based on an adaptive Model Predictive Control and PID control that utilizes selected AI methods. The presented approach presents an updated the mathematical model of an Eddy Current dynamometer based on experimentally obtained system operational data. Moreover, the proposed PID control system proves changeable Kp (Proportional gain), Ki, (Integral gain) and Kd (Derivative gain) parameters and MPC with continuously-updated the mathematical model of an Eddy Current Dynamometer in real-time to adapt and presents a good capacity to adapt nonlinearities and bring robustness. Finally, the comparison of AI methods and related learning performances based on the assessment technique of mean absolute percentage error (MAPE) issues are discussed. The results indicate that basic feed-forward Neural Network, General Regression Neural Network (GRNN), Radial Basis Network (RB), Neuro Fuzzy Network (ANFIS) coupled MPC and PID have quite satisfying performances.
Experimental study of active flow control with synthetic jet vortex generators on airfoil aerodynamics
This study investigates the active flow control of a selected airfoil type to improve aerodynamic performance. Synthetic jet actuators have demonstrated promising results for active flow control with various orifice structures in aviation systems. In the first part of the study, jet structures formed at different orifice exits are examined using synthetic jet actuators for flow control. In the second part, the aerodynamic performance of the NACA0020 airfoil with different orifice structures is evaluated using synthetic jet actuators. Experimental results from the first part reveal the jet structures formed at the exits of different aspect ratio (AR) orifices and models using synthetic jet actuators. The distribution patterns of these jet structures in various models and geometric ratios are analyzed using the oil visualization method. It is observed that different orifice structures influenced jet formation differently. Additionally, variations in frequency and voltage impacts vortex structures. In the second part, the experimental results indicate that synthetic jet actuators placed on the suction surface of the main profile reduce drag, increase lift, shift the stall angle, and enhance the overall performance of the wing profiles.