ANALYSIS OF WELDING PARAMETERS IN GAS METAL ARC WELDING BY A WELDING ROBOT Erener, Yavuz M.S., Department of Mechanical Engineering Supervisor : Prof. Dr. R. Tuna Balkan Co-Supervisor : Prof. Dr. M. A. Sahir Arikan September 2006, 130 pages In Robotic Gas Metal Arc Welding process, the welding parameters controlled by the welder (travel speed of the welding torch, wire feed speed, current, voltage, wire diameter, etc.) should be considered to obtain a desired welding quality. To design an appropriate welding model for the used equipment, the effects of each parameter should be studied by carrying out an adequate number of experiments. The welding process is described by analyzing the experimental data to define the relationships between the welding parameters and process variables. Various regressional models can be suggested to establish the analytical relationships. In this study, the relationship between bead geometry and voltage, current, travel speed and wire feed speed is established by using a specific computer program developed for this purpose.

In this study, a design procedure incorporating a genetic algorithm (GA) is developed for optimization of structures. The objective function considered is the total weight of the structure. The objective function is minimized subjected to displacement and strength requirements. In order to evaluate the design constraints, finite element analysis are performed either by using conventional finite element solvers (i.e. MSC/NASTRAN&reg / ) or by using in-house codes. The application of the algorithm is shown by a number of design examples. Several strategies for reproduction, mutation and crossover are tested. Several conclusions drawn from the research results are presented.

The aim of this thesis is to develop a test facility in order to validate the safe separation of a store model from the aircraft model with the analysis of the acquired data. Different alternatives are considered in the conceptual design and a facility is selected which contains the rail constructed 5 meters above the ground. 300 meter test range is used to accelerate the aircraft model, attached to the carriage with solid rocket engines. The concept is to accelerate the carriage within the first 100 meters, to perform the separation of the model at the desired velocity when the rocket engine is burned out, and in the last 100 meters, to decelerate the carriage in order to recover the models. After the parts of the carriage are designed with respect to the requirements of the separation and completing the similitude study of scaled models with the ratio of 1/8, separation tests are performed. These tests are constrained with 0.9 Mach due to the store and test aircraft limitations. Doppler radar is issued to measure the velocity and photogrammetry method is used to validate and analyze the separation and to obtain the vertical trajectory of the store model. Measured trajectory of the store model is compared with the trajectory of the full scale store.

In this study, an attempt is made to obtain convergent and stable solutions of the K-E turbulence model equations for non-reacting turbulent flows over an isothermal solid surface in regression. A physics based mathematical model is used to describe the flow and temperature field over the moving surface. The flow is assumed to be two-dimensional, unsteady, incompressible with boundary layer approximations. Parabolized form of the standard K-E equations is adopted to simulate turbulence in the flow. Regression of the solid surface causes the bounds of the solution domain to change with time, therefore a coordinate transformation is used in the vertical direction. The computational domain with fixed boundaries is discretized using an orthogonal grid system where a coordinate stretching is used in the vertical direction. A second order accurate, explicit finite difference technique is used for discretization of the governing equations. The final set of discretized equations is then solved using a solution algorithm specifically developed for this study. The verification of the solution algorithm includes a grid independence study, time increment study, and a comparison of the steady state results for the laminar and the turbulent flow cases. Finally, a parametric study is conducted using the proposed solution algorithm to test the stability of the numerical results for different Reynolds numbers, regression rates, and surface temperatures. It is concluded that the proposed numerical solution algorithm is capable of providing convergent and stable solutions of the two-equation turbulence model.

In this study, an optimization code has been developed based on genetic algorithms associated with the finite element modeling for the shape optimization of plane stress problems. In genetic algorithms, constraints are mostly handled by using the concept of penalty functions, which penalize infeasible solutions by reducing their fitness values in proportion to the degrees of constraint violation. In this study, An Improved GA Penalty Scheme is used. The proposed method gives information about unfeasible individual fitness as near as possible to the feasible region in the evaluation function. The objective function in this study is the area of the structure. The area is minimized considering the Von-Misses stress criteria. In order to minimize the objective function, one-point crossover with roulette-wheel selection approach is used. Optimum dimensions of four problems available in the literature have been solved by the code developed . The algorithm is tested using several strategies such as / different initial population number, different probability of mutation and crossover. The results are compared with the ones in literature and conclusions are driven accordingly.

Fluidized bed is a suitable technology for combustion and gasification of biomass materials. Hydrodynamics occurring in the bed is crucial for the design and operation of the combustion or gasification unit. In the present study, hydrodynamic behavior of binary mixtures of biomass-silica sand in a bubbling fluidized bed was experimentally investigated. Five different biomass materials and silica sand with three different particle sizes were employed to form binary mixtures. Biomass materials were rice husk, sawdust, wheat straw, hazelnut shell and olive cake which are all potential energy sources for Turkey. Effects of mass percentage of biomass and particle size of silica sand on minimum fluidization velocity of the mixtures were determined. Comparisons between results of the present study and predictions of available correlations proposed for minimum fluidization velocity of binary mixtures were carried out. Mixing and segregation characteristics of biomass-silica sand binary mixtures were investigated for mixtures having different mass fraction of biomass and different silica sand particle sizes. Fluidization and bubbling behaviors of mentioned mixtures were observed in a 2-D fluidized bed and images taken during steady-state operation of bed were presented as visual tools to guide fluidization characteristics of the bed. Mass percentage increase of rice husk, wheat straw and sawdust resulted in increase in minimum fluidization velocity of the mixture whereas change in mass fraction of olive cake and hazelnut shell had no effect on minimum fluidization velocity. Minimum fluidization velocity increased with increase of silica sand particle size for all biomass-silica sand mixtures having same mass percentage of biomass. Vertical mixing pattern in the bed at steady state conditions were found almost same for all biomass-silica sand mixtures. Biomass acted as flotsam and accumulated mostly at the top of the bed and silica sand acted as jetsam and accumulated mostly at the bottom of the bed. 2-D bed experiments showed that mixing biomass materials with silica sand provides desired bubbling behavior in the bed.

In this thesis, load scheduling problem for cogeneration plants is interpreted in the context of stochastic programming. Cogeneration (CHP) is an important technology in energy supply of many countries. Cogeneration plants are designed and operated to cover the requested time varying demands in heat and power. Load scheduling of cogeneration plants represents a multidimensional optimization problem, where heat and electricity demands, operational parameters and associated costs exhibit uncertain behavior. Cogeneration plants are characterized by their &lsquo / heat to power ratio&rsquo / . This ratio determines the operating conditions of the plant. However, this ratio may vary in order to adapt to the physical and economical changes in power and to the meteorological conditions. Employing reliable optimization models to enhance short term scheduling capabilities for cogeneration systems is an important research area. The optimal load plan is targeted by achieving maximum revenue for cogeneration plants. Revenue is defined for the purpose of the study as the sales revenues minus total cost associated with the plant operation. The optimization problem, which aims to maximize the revenue, is modeled by thermodynamic analyses. In this context, the study introduces two objective functions: energy based optimization, exergy-costing based optimization. A new method of stochastic programming is developed. This method combines dynamic programming and genetic algorithm techniques in order to improve computational efficiency. Probability density function estimation method is introduced to determine probability density functions of heat demand and electricity price for each time interval in the planning horizon. A neural network model is developed for this purpose to obtain the probabilistic data for effective representation of the random variables. In this study, thermal design optimization for cogeneration plants is also investigated with particular focus on the heat storage volume.

Due to the high performance of electronic components, the heat generation is increasing dramatically. Heat dissipation becomes a significant issue in efficiency promotion and stable operation. Microchannels are of current interest for use in heat exchangers where very high heat transfer performance is desired. Microchannels provide high heat transfer coefficients because of their small hydraulic diameters. In this study, the design and experimental investigation of fluid flow and heat transfer in a microchannel heat exchanger is conducted. Water and air are used as the working fluids and flowed through microchannels. The heat exchanger has been designed with 6 rows of microchannels for water flow and 7 rows of microchannels for forced flow of air. The heights of the microchannels are 4 mm and 10 mm respectively for water and air flows. Microchannels are brazed to form the heat exchanger. For forced convection cooling with air, a military fan is used. A constant heat source has been specifically designed for experiments. Water flow and heat transfer experiments are conducted on the aluminum microchannel heat exchanger. An experimental method of imposing a constant heat flux to water prior to the entrance to the microchannel heat exchanger, to adjust the inlet temperatures is used. v From the data obtained, the rate of heat transfer, effectiveness and various other parameters have been computed and the results have been compared with those from an available commercial heat exchanger. The results indicate that the heat exchanger performs well and provides 681 W of cooling in a volume 677.6 cm3 while the commercial heat exchanger provides 702.5 W of cooling in a volume 2507.5 cm3. In addition, air-side Colburn modulus has been obtained with respect to Reynolds number.

An experimental study has been performed to investigate the condensation on the inner surface of open cavities located on horizontal and vertical surfaces of a refrigerator to simulate ice dispensers, water dispensers, electronic display slots, and door handles, etc. Cavity size, cavity depth and insulation thickness were variable parameters. Cavities were located on fresh food and freezer compartments to study two different boundary conditions. Level of condensation was put into a scale based on visual observation for condensed water droplets on the cavity surface. As a practical tool for design engineers, insulation thickness was plotted as a function of cavity depth indicating the level of condensation. Similar cavity geometries were tested on both freezer and fresh food compartments. Aluminum foil attached behind the inner surface of a cavity located on the fresh food door helped to decrease the level of condensation significantly. However, condensation can not be avoided for the cavities located on the freezer compartment deeper than critical values. Forming radius at the corners and on the edges of cavities decreased condensation on these regions. The effects of air circulation on condensation inside and outside the refrigerator were also investigated for some cavities. Particle Image Velocimetry (PIV) method was used to visualize non-disturbed air flow pattern over the cavity and at the cavity opening. High convective heat transfer at the cavity center was strongly associated with relatively high air velocity. Temperature distribution and flow pattern were analyzed by a CFD Programme. Condensation due to relatively low temperatures at the corners and on the edges was associated with conductive heat transfer in multiple directions and insufficient air circulation on these regions.

In dimensional measurements, it is required to specify uncertainty in the measurement as the range of possible deviation for the measurement result. In this thesis, a computer program is developed for evaluating uncertainty in measurement of commonly used dimensional measuring devices like vernier callipers, micrometers, comparators, and gauge blocks. In evaluation of the uncertainty in measurement, some uncertainty sources like temperature difference between the measured part and the instrument, uncertainty in reference gauge block&rsquo / s dimension, mechanical effects, etc. are considered. The program developed, employs the EAL, NIST and GUM uncertainty evaluation equations as standard equations. However, the program can also be used for other measuring instruments and the users can define their own uncertainty equation. In the evaluations, for the standard uncertainty of the variables considered, symmetric distributions are used. The program gives the uncertainty budget and to compare the contribution of each variable on the overall uncertainty of the measurement, the uncertainty effect ratio is also given. In this thesis the evaluation process for uncertainty in measurement, the difference between the measurement error and uncertainty in measurement and the structure of the program are discussed. Also, a set of experiments has been made to illustrate the application of the program for evaluating the measurement uncertainty of vernier callipers with 1/50 and 1/20 resolutions, digital vernier calliper and 25 mm micrometer.