The thermal control of a die is crucial for the development of high efficiency injection moulds. For successful thermal management, this research provides an effective control strategy to find sensor locations, identify thermal dynamic models, and design controllers. By applying a clustering method and sensitivity analysis, sensor locations are identified. The neural network and finite element analysis techniques enable the modeling to deal with various cycle-times for the moulding process and uncertain dynamics of a die. A combination of off-line training through finite element analysis and training using on-line learning algorithms and experimental data is used for the system identification. Based on the system identification which is experimentally validated using a real system, controllers are designed using fuzzy-logic and self-adaptive PID methods with backpropagation (BP) and radial basis function (RBF) neural networks to tune control parameters. Direct adaptive inverse control and additive feedforward control by adding direct adaptive inverse control to self-adaptive PID controllers are also provided. Through a comparative study, each controller’s performance is verified in terms of response time and tracking accuracy under different moulding processes with multiple cycle-times. Additionally, the improved cooling effectiveness of the conformal cooling channel designed in this study is presented by comparing with a conventional straight channel.

Cold metal forming processes have been widely used for manufacturing of their high production rates and increased yield strength after forming process. For the use in service, increased yield strength of the cold-formed products should be known. The new yield strength can be found by several methods. Mechanical tests such as compression or tensile test are direct methods to obtain new yield strength if the product shape is appropriate. Finite element simulations may be another way to get accurate results for new yield strength distribution. Also Vickers hardness number can be used for prediction of yield strengths by available conversion models. The aim of this study is to compare the results of all these methods. During the study two different materials (austenitic stainless steel and carbon steel) cold formed by drawing and extrusion are investigated. FE simulations have been conducted to predict product properties. For this purpose flow curves obtained from compression and tensile tests are used in FE-models based on elasto-plastic, isotropic hardening material. Results show that both materials are highly anisotropic and have much lower yield strength values than found in simulations. Similarly none of the models correlating Vickers hardness numbers and yield strengths are successful since they are designed for an isotropic hardening material. This study basically presents the deviation of a real material behavior from isotropic material behavior.

Producing axi-symmetrical parts with holes from tubular stock by tube upsetting is a frequently used technique in industry. There are basically four types of tube upsetting process / external, internal, simultaneous internal and external upsetting, and expanding of tube. In general, tubular parts require more than one upsetting stage. In industry, generally trial-error methods, which require lots of time and effort depending on experience, are used for the design of stages. Wrong design causes failures during production. On the other hand, the problems, which are likely to be encountered in manufacturing, can be observed and solved in the design stage by using finite element analysis. In this study, the finite element analyses of external, internal, simultaneous internal and external tube upsetting, and tube expanding processes have been realized. During the analyses, the part and the die geometries at the intermediate stages, which have been designed according to the proposed procedures, have been used. The stress and strain distributions and die filling actions have been observed during the process. The process design and die geometries have been evaluated according to the finite element results. It has been seen that the recommended procedures generally generate acceptable designs. In some cases, it has been noted that minor modifications may be required on the design.

In this thesis a 3-D Euler flow solver is coupled with a finite element program in order to solve static aeroelastic problems involving aircraft wings. A loosely coupled solution approach based on an iterative solution procedure is used to solve the coupled field problem. Because of the deformation of the underlying surface over which the flow is solved, Computational Fluid Dynamics mesh has to move at each computational aeroelastic iteration in order to comform to the new shape of the aerodynamic surface. As a part of this work, a procedure is developed in order to move fluid grid points, which views the whole computational domain as an isotropic elastic medium and solves it using finite element method. A matching discrete interface is defined / displacement and pressure data exchange is accomplished at this interface. AGARD Wing 445.6 and an elastic supercritical wing is modelled and solved with the developed computational aeroelastic procedure and the obtained results are compared with numerical and wind tunnel data.

In this study settlement behaviour of K&uuml / rt&uuml / n dam, which is the first concrete faced rockfill dam in Turkey, is investigated. Two dimensional plane strain finite element analyses are carried out in order to determine the total stresses and displacements during construction and reservoir filling conditions. Hardening soil model is used in order to represent the non-linear, inelastic and stress dependent behaviour of rockfill material. Material model parameters are selected mainly referring to the previous studies on the dams consisting of similar materials. Calculated stresses and settlements are compared with the observed values and in general, they were found to be in good agreement for the construction stages. It is seen that, due to the relatively narrow valley and steep abutment slopes, arching is a significant parameter as far as the stresses and settlements are concerned. For the reservoir impounding condition, calculated settlements were found to be slightly larger than the observed values, which may indicate that during the reservoir impounding, the rockfill embankment behaves in a stiffer manner as compared to that of during construction stages.

Sheet metal forming processes, especially deep drawing processes give diverse results by various materials. Extreme differences occur between steel sheets and aluminum sheets. The main causes of this variance are anisotropy, elastic modulus and microscopic material properties. The aim of this thesis is to evaluate the deep drawing properties and also to develop suitable process parameters for aluminum and steel sheets by finite element simulation. In the simulation, the commercial dynamic-explicit code PAM-STAMP has been used. The reliability of the finite element package was verified by a comparison with the NUMISHEET 2002 benchmarks. Additionally, a commercial part is numerically simulated for experimental verification. The results of the simulations have been compared with several experiments that were performed in Metallurgical and Materials Engineering and Mechanical Engineering Departments. Finally, the simulation results are compared with analytical expressions for verification of results. The materials investigated for the deep drawability comparison is a deep drawing quality mild steel and an aluminum alloy designated as 6111-T4. For experimental verification St4 steel is used. Results are in agreement with the fact that aluminum and steel materials behave differently upon deep drawing in terms of the onset of failure, wrinkling and final shape. Aluminum is found to be less formable than steel for cup drawing operations.

This dissertation research focuses on micromechanical modeling and simulations of two-phase heterogeneous materials exhibiting anisotropic and non-uniform microstructures with long-range spatial correlations. Completed work involves development of methodologies for realistic micromechanical analyses of materials using a combination of stereological techniques, two- and three-dimensional digital image processing, and finite element based modeling tools. The methodologies are developed via its applications to two technologically important material systems, namely, discontinuously reinforced aluminum composites containing silicon carbide particles as reinforcement, and boron modified titanium alloys containing in situ formed titanium boride whiskers. Microstructural attributes such as the shape, size, volume fraction, and spatial distribution of the reinforcement phase in these materials were incorporated in the models without any simplifying assumptions. Instrumented indentation was used to determine the constitutive properties of individual microstructural phases. Micromechanical analyses were performed using realistic 2D and 3D models and the results were compared with experimental data. Results indicated that 2D models fail to capture the deformation behavior of these materials and 3D analyses are required for realistic simulations. The effect of clustering of silicon carbide particles and associated porosity on the mechanical response of discontinuously reinforced aluminum composites was investigated using 3D models. Parametric studies were carried out using computer simulated microstructures incorporating realistic microstructural attributes. The intrinsic merit of this research is the development and integration of the required enabling techniques and methodologies for representation, modeling, and simulations of complex geometry of microstructures in two- and three-dimensional space facilitating better understanding of the effects of microstructural geometry on the mechanical behavior of materials.

The objective of this study is to formulate a general 3D material-structural analysis framework for the thermomechanical behavior of steel-concrete structures in a fire environment. The proposed analysis framework consists of three modeling parts: fire dynamics simulation, heat transfer analysis, and a thermomechanical stress analysis of the structure. The first modeling part consists of applying the NIST (National Institute of Standards and Technology) fire dynamics simulator (FDS) where coupled Computational Fluid Dynamics (CFD) with thermodynamics are combined to model the fire progression within the steel-concrete structure. The goal is to generate the spatial-temporal (ST) solution variables (temperature, heat flux) on the surfaces of the structure. The FDS-ST solutions are generated in a discrete numerical form. Continuous FDS-ST approximations are then developed to represent the temperature or heat-flux at any given time or point within the structure. An extensive numerical study is carried out to examine the best ST approximation functions that strike a balance between accuracy and simplicity. The second modeling part consists of a finite-element (FE) transient heat analysis of the structure using the continuous FDS-ST surface variables as prescribed thermal boundary conditions. The third modeling part is a thermomechanical FE structural analysis using both nonlinear material and geometry. The temperature history from the second modeling part is used at all nodal points. The ABAQUS FE code is used with newly developed external user subroutines for the second and third simulation parts. The main objective is to describe the nonlinear temperature-dependency of the specific heat of concrete materials, especially high-strength concretes, that drastically affects their transient thermal solution. New algorithms are also developed to apply the continuous FDS-ST surface nodal boundary conditions in the transient heat FE analysis. The proposed modeling framework is applied to predict the temperature and deflection of the well-documented Cardington fire tests and to predict the time-to-collapse of the recent Oakland bridge fire caused by a fuel-truck accident.

Frontal Protection Systems (FPS) have become a popular accessory for passenger vehicles. They are used to protect the front of a vehicle during minor impacts, and to attenuate the impact energy during major impacts. With the increased safety of modern passenger vehicles, the fitment of a FPS to a vehicle requires careful consideration to the design and installation of the FPS as they may modify vehicle crush characteristics. This is particularly important in vehicles fitted with air bags. These community and industry concerns triggered the research discussed in this thesis, which is the first comprehensive project undertaken in this particular area. This project generated comprehensive research knowledge on the impact response and energy absorption of FPS in order to evaluate performance. This involved a range of experimental testing supplemented by finite element analysis. Experimental testing was conducted using quasi-static and dynamic techniques to assess the overall performance of current FPS available. Finite element models were then generated and analysed using both implicit and explicit techniques, and calibrated against the experimental testing results. These models were used throughout the project to assess the FPS response, in particular the energy absorbed, to changes in impact characteristics. FPS assessment guidelines were developed from the knowledge generated from the numerous FPS tests and analyses carried out in this research project. These guidelines have been used in the design and evaluation of a number of FPS for airbag compatibility. The real life performance of vehicles fitted with these FPS, have given confidence to the assessment criteria developed in this research project. This project has demonstrated that FPS can be designed to complement the safety systems of modern passenger vehicles, and thus passenger safety. This would not have been possible without the comprehensive research carried out in this project.

Thesis (Doctora).