Dissimilar-metal welds(DMWs) are used widely in various industrial applications due to the pratical importance from the technical and economic aspect. However, DMWs have several fabricative and metallurgical drawbacks that can often lead to in-service failures. For example, the most pronounced fabrication faults are hot cracks. Recently, DMWs have used the various of heat source to decrease such as faults. In this paper, the weldability on dissimilar metal welds of Austenite stainless steel and Low carbon steel using a continuous wave Nd:YAG laser was experimentally investigated. Experiments were conducted to determine effects of welding parameters, on eliminating or reducing the extent welding zone formation at dissimilar metal welds and to optimize those parameters that have the most influence parameters such as focus length, power, beam speed, shielding gas, and wave length of laser.

Polishing pads play an important role in chemical mechanical polishing(CMP) which has recently been recognized at the most effective method to achieve global planarization. In this paper, we have investigated CMP characteristics as a change of groove density of polishing pads. The parameter(K_n) is proposed to estimate groove density of pad. The K_n is defined as groove area divided by pitch area. As the groove density value increased, removal rate increased to some point and then gradually saturated in case of increasing the groove density excessively. In addition Within wafer non-uniformity(WIWNU) worse as groove density increased excessively, although WIWNU improved as groove density increased. Also the uniformity of temperature of pad surface decreased as the groove density increased. It was because that the cooling effect increased as groove density increased. In other words, increasing the groove density which means the apparent contact area of pad has influence on amount of discharge of slurry during polishing process.

Generally, it is not easy to get a suitable controller for multi variable systems. If the modeling equation of the system can be found, it is possible to get LQR controller as an optimal solution. This paper suggests an LQR learning method tc design LQR controller without the modeling equation. The proposed algorithm uses the same cost function with errol and input energy as LQR is used, and the LQR controller is trained to reduce the function. In this training process, the; Jacobian matrix that informs the converging direction of the controller is used. Jacobian means the relationship of output variations for input variations and can be approximately found by the simple experiments. In the simulations of 2 hydrofoil catamaran with multi variables, it can be confirmed that the training of LQR controller is possible by using the; approximate Jacobian matrix instead of the modeling equation and this controller is not worse than the traditional LQR controller.

Our purpose is to develop a precision magnetic displacement sensor that has sub-micron resolution and small size probe. To achieve this, we first have tried to establish mathematical models of a magnetic sensor in this paper. The inductance model that presents basic measuring principle of a magnetic sensor is based on equivalent magnetic circuit method. Especially we have concentrated on modeling of magnetic flux leakage and magnetic flux fringing. The induced model is verified by experimental results. The model, including the magnetic flux leakage and flux fringing effects, is in good agreement with the experimental data. Subsequently, based on the augmented model, we will design magnetic sensor probe in order to obtain high performances and to scale down the probe.

This paper is described a development of an optical fiber displacement sensor. The optical fiber sensor using an intensity modulated measures the displacement between target and sensor. A prototype sensor is composed of a transmitting part, a receiving part and a signal processing circuit. The experiment was conducted not only the sensor performance but also factors that affect intensity. The main performance of this sensor is resolution of 0.37 ㎛ and the non-linearity 0.7％ FS and the dynamic bandwidth of about 6.3 kHz. As a result of rotation test, the prototype sensor showed an equivalent performance to a commercial eddy current sensor.

Development of a locomotive mechanism for the capsule type endoscopes will largely enhance the ability to diagnose disease of digestive organs. In connection with it, most of researches have focused on an installable locomotive mechanism in the capsule. In this paper, it is introduced that the movement of a capsule type endoscope in digestive organ can be manipulated by magnetic force produced outside human body. Since the magnetic force is provided by permanent magnets, no additional power supply to the capsule is required. Using a robotic manipulator for locating the external magnet, the capsule motion control system can cover the whole human digestive organs. This study is particularly concentrated on dither motion effect to improve the mobility of capsule type endoscope. It was experimentally found out that the friction coefficient between the capsule and digestive organ can be remarkably reduced by superposing yawing or rolling dither motion on the translatory motion. In this paper, the experimental results obtained with the direction, amplitude and frequency of sinusoidal dither motion changed is reported.

The development of an accurate and energy saving pneumatic regulator that may be applied to a variety of practical pressure control applications is described in this paper. A novel modified pulse width modulation(MPWM) valve pulsing algorithm allows the pneumatic regulator to become energy saving system. A comparison between the system response of conventional PWM algorithm and that of the modified PWM(MPWM) algorithm shows that control performance is almost the same, but energy saving is greatly improved by adopting this new MPWM algorithm. The effectiveness of the proposed control algorithm are demonstrated through experiments with various reference trajectories.

This study has been focused on the development of automatic bottle air rinser. It is designed to rinse clean empty bottles prior to filling. The bottles are automatically indexed beneath an air nozzle that has both a clean airjet and vacuum source. The bottle is first given a burst of clean air to loosen any particles from the wall of the bottle. A vacuum sequence follows which removes all particulates into a self contained filter unit. In order to the provide the desired function, analysis is carried out by FEM simulation using FLUENT and CATIA software. The final results of analysis are applied to the design of automatic bottle rinser and the machine is successfully developed.

In this study a general method for computing offsets of free form curves is presented. In the method arbitrary free form curve is approximated with point series considering required tolerance. The point series are offset precisely using the normal vectors computed at each point and loop removal is carried out by the newly suggested algorithm. The resulting offset points are transformed to lines and arcs using the biarc approximation method. Tangent vectors for approximation of discrete points data are calculated by traditional local interpolation scheme. In order to show the validity and generality of the proposed method , various offsettings are carried our for the base curves with complex shapes.

The role of microstructure is quite significant in fretting of Ti-6AI-4V since its material properties depend strongly on crystallographic texture. In this study, we adopt crystal plasticity theory with a 2-D planar triple slip idealization to account for microstructure effects such as grain orientation distribution, grain geometry, as well as a colony size. Crystal plasticity simulations suggest strong implications of microstructure effects at fretting contacts.

Tungsten heavy metal is characterized by a high density and novel combination of strength and ductility. Among them, 90W-7Ni-3Fe is used for applications, where the high specific weight of the material plays an important role. They are used as counterweights, rotating inertia members, as well as for defense purposes(kinetic energy penetrators, etc.). Because of these applications, it is essential to detemine the dynamic characteristics of tungsten alloy. In this paper, Explicit FEM(finite element method) is employed to investigate the dynamic characteristics of tungsten heavy metal under base of stress wave propagation theory for SHPB, and the model of specimen is divided into two parts to understand the phenomenon that stress wave penetrates through each tungsten base and matrix. This simulation results were compared to experimental one and through this program, the dynamic stress-strain curve of tungsten heavy metal can be obtained using quasi static stress-strain curve of pure tungsten and matrix.

The objective of the study is to improve the quality of wheel bearing hub by the rigid-plastic finite element analysis and the response surface methodology. The rigid-plastic finite element codes, AFDEX-2D and DEFORM-3D, were used to analyze the two-dimensional and three-dimensional forging processes, respectively. The response surface analysis is used to find the minimum underfill by the variation of design variables such as the height of billet after upsetting and punch angles of blocker dies. The metal flow of forged product shows good agreement with the results from 20 and 3D analysis. Also, the quality of the wheel bearing hub has been improved by the optimization of design variables and the machining time has been reduced by the machining allowance.

In modem industrial society, many kinds of valves have been used to control fluid flow. For the optimization of a ball valve, a quarter of spherical ball was modeled and evaluated. However, there is a difficulty in the application of the optimization technique because of the complexity of the fluid and the restriction of analysis.
Therefore, in this study, it is performed to evaluate the stability of an initial model using FEM(Finite Element Method) and CFD. In addition, a shape optimization design of the valve is accomplished using an orthogonal arrangement and characteristic functions. From the result, a new design method is represented that could overcome the time and space restriction in structural design, such as the divided ball valve with less experiment.

A spatial 3 degrees-of-freedom mechanism employing Stewart Platform structure is proposed: the mechanism maintains the 3 - RRPS structure of Stewart Platform but has an additional passive PRR serial sub-chain at the center area of the mechanism in order to constrain the output motion of the mechanism within the output motion space of the added P RR serial subchain. The forward and reverse position analyses of the mechanism are performed. Then the mechanism having both the forward and the reverse closed-form solutions is suggested and its closed form solutions are derived. It is confirmed, through the kinematic analysis of those two proposed mechanisms via kinematic isotropic index, that both the proposed mechanisms have fairly good kinematic characteristics compared to the existing spatial 3-DOF mechanisms in literature ^1.6.