Sehun Rhee

Modelling and optimization of a GMA welding process by genetic algorithm and response surface methodology

The welding process, due to its complexity, has relied on empirical and experimental data to determine its welding conditions. However, trial-and-error methods to determine optimal conditions incur considerable time and cost. In order to overcome these problems, a genetic algorithm and response surface methodology have been suggested for determining optimal welding conditions. First, in a relatively broad region, near-optimal conditions were determined through a genetic algorithm. Then, the optimal conditions for welding were determined over a relatively small region around these near-optimal conditions by using response surface methodology. In order to give different objective function values according to the positive or negative response from the set target value in the optimization problem, a desirability function approach was used. Application of the method proposed in this paper revealed a good result for finding the optimal welding conditions in the gas metal arc (GMA) welding process.

New technology for measuring dynamic resistance and estimating strength in resistance spot welding

Because welded structures such as automotive bodies have become smaller and lighter, a strong emphasis is being placed on the quality of the welds. Over the years many quality estimation systems have been developed by several researchers in order to ensure that welds are of high quality. The dynamic resistance signal, which is closely related to the nugget formation of the weld, has been used very effectively for this purpose, together with the electrode displacement signal. However, in previous research, the dynamic resistance was calculated using process parameters that were measured in the secondary circuit of the welding machine, causing many in-process problems. In this study, the process variables, which were monitored in the primary circuit of the welding machine, are used to obtain the variation of the dynamic resistance across electrodes. This allows the dynamic resistance monitoring system to be applied to the in-process system without any extra monitoring devices in the secondary circuit. Also, in order to test the reliability of such a system, an artificial intelligence algorithm to estimate the weld quality using the primary dynamic resistance is proposed.

Real time estimation of CO2 laser weld quality for automotive industry

Laser welding is one of the most precise welding processes in joining sheet metals. In laser welding, performing real time evaluation of the welding quality is very important to enhance the efficiency of the welding process. In this study, the plasma and spatter, which are generated during laser welding, are measured using UV and IR photodiodes. The factors that influence weld quality are classified into five categories; optimal heat input, slightly low heat input, low heat input, partial joining due to gap mismatch, and nozzle deviation. The data number deviated from reference signals and their standard deviations were also considered to evaluate the qualities. A system was also formulated to perform real time evaluations of the weld quality using a fuzzy multi-feature pattern recognition with the measured signals.

Estimation of weld bead size in CO2 laser welding by using multiple regression and neural network

On the laser weld production line, a slight alteration of welding condition changes the bead size and the strength of the weldment. A measurement system is produced by using three photodiodes for detection of the plasma and spatter signal in CO2 laser welding. The relationship between the plasma or spatter and the bead shape, and the mechanism of the plasma and spatter signals were analyzed for the bead size estimation. Penetration depth and bead width were estimated using multiple regression analysis and the neural network.

Optimization of welding parameters for resistance spot welding of TRIP steel with response surface methodology

Many automotive companies are endeavouring to reduce the weight of the car body in response to various environmental issues. One initiative is the development of TRIP (Transformation Induced Plasticity) steels with a high strength and ductility. Resistance spot welding is a complex process, which requires specific optimal welding conditions based on experimental data. However, the trial-and-error method to determine the optimal conditions requires a large number of experiments, and so response surface methodology has been employed to overcome this problem. The second-order model was used here. This has been used in the resistance spot welding process of the TRIP steel and galvanized TRIP steel with a zinc-coated layer to optimize the welding parameters. The welding current, welding time, and welding force were selected as input variables, and the shear strength and indentation were selected as output variables.

A fuzzy pattern recognition based system for monitoring laser weld quality

In-process monitoring of welding has become important as the use of laser welding increases. Plasma and spatter are measured and used as signals for estimating the quality of a weld. The measurement system consists of three photodiode sensors (one IR and two UV) to detect the plasma and spatter signals in CO2 laser welding. The estimating algorithm was constructed using fuzzy pattern recognition considering the amplitudes as well as amounts of data beyond the tolerance boundary. Weld qualities were classified as optimal heat input, slightly low heat input, low heat input and misalignment of focus. Also, an algorithm for detecting spatter was created in order to find the partially produced pit. These algorithms were used for quality monitoring in tailored blank welds with a CO2 laser.

Design and fabrication of a six-component force/moment sensor

Abstract This paper describes the development of a six-component force/moment sensor with the plate-beams which may be used in industry for measuring forces F x , F y , F z and moments M x , M y , M z simultaneously. In order to device the six-component force/moment sensor, the procedures are performed as follows: first, the equations to predict the bending strains on the surfaces of the plate-beams under the forces or the moments are derived, second, the sizes of the sensing elements of the force/moment sensor are determined by using the derived equations, third, the Finite Element Method (FEM) analysis for confirming the strains from the theoretical analysis is performed, next, the attachment locations of the strain gages of each sensor are selected, then, the six-component force/moment sensor is fabricated, and lastly, a characteristic test of the six-component force/moment sensor is performed. It reveals that the rated strains calculated from the derived equations are in agreement with the results from the FEM analysis and the experiments. The interference errors of each sensor are less than 2.0%.

Development of an arc sensor model using a fuzzy controller in gas metal arc welding

The modelling of an arc sensor and the determination of control variables play a key role in applying the fuzzy controller to a seam tracking system that uses an arc sensor. In this study, experimental modelling was performed on an arc sensor that uses the current area difference. This is a sensory method that is based on the dynamic behaviours of the arc current. Also, the scaling factors and control variables of the fuzzy controller were determined by way of simulated experiments. The experimental model was comprised of a regression model, expressing the relationship between the current area difference and offset distance through weaving experiments, and a noise term, which uses the standard deviation. Based on this model, a seam tracking simulation system was formulated, and used to determine the optimal scaling factors of the fuzzy controller. In an actual seam tracking experiment, using the optimal scaling factors acquired through the simulation, the suggested seam tracking fuzzy controller showed satisfactory performance.

Design and fabrication of a three-component force/moment sensor using plate-beams

This paper describes the development of a three-component force/moment sensor with plate-beams which may be used for measuring the forces Fx, Fy and the moment Mz simultaneously in industry. In order to make the three-component force/moment sensor, the following procedures are performed. (1) Derivation of equations to predict the bending strains on the surfaces of the plate-beams under the forces or the moments. (2) Determination of the sizes of the sensing elements of the force/moment sensor using the derived equations. (3) Finite Element Method (FEM) analysis for confirming the strains from the theory. (4) Selection of the attachment locations of the strain gauges of each sensor. (5) Fabrication of the three-component force/moment sensor. (6) Characteristic test of the three-component force/moment sensor. The rated strains calculated from the derived equations agree well with the results from the FEM analysis and the experiments. The interference errors of each sensor are less than 1.7%.

Development of a weld quality monitoring system in CO2 laser welding by using photodiodes

On a laser weld production line, a slight alteration of the welding condition produces many defects. The defects are monitored in real time in order to prevent continuous occurrence of defects, to reduce the loss of material, and to guarantee good quality. The measurement system consists of three photodiodes to detect the plasma and spatter signals in CO2 laser welding. For high speed CO2 laser welding, such as laser tailored welded blanks, an on-line weld quality monitoring system was developed by using fuzzy multifeature pattern recognition. Weld qualities are classified as optimal heat input, little low heat input, low heat input, and focus misalignment. Then the result of the final weld quality is estimated good or bad.