This study proposed a conditional generative adversarial network (cGAN) model for predicting steel plate deformation based on heating line positions in a line heating process. A database was constructed by performing finite element analysis (FEA) to establish relationships between heating line positions and deformation shapes. Deformation shapes were converted into color map images. Heating line positions were used as conditional labels for training and validating the proposed model.
During the training process, generator and discriminator loss values, along with MSE and R² metrics, converged stably, demonstrating that generated images closely resembled the actual data. Validation results showed that predicted deformation magnitudes had an average relative error of approximately 3% and a maximum error of less than 7%. These findings confirm that the proposed model can effectively predict steel plate deformation shapes based on heating line positions in the line heating process, making it a reliable predictive tool for this application.

When a narrow gap was formed under appropriate welding conditions in the steel pipe manufacturing process using highfrequency resistance welding, temperature distribution was analyzed to predict the length of the gap. Assuming the length of the gap from the apex point to the welding point at an applied voltage, and calculating the temperature distribution around the gap, the length of the gap with an appropriate fusion width at the welding point could be estimated. Along with this, the current density and magnetic flux density distributions that appeared in the narrow gap were obtained according to the change in the applied voltage, and the distribution shape and size of the electromagnetic force acting on the gap were also predicted. The current density, magnetic flux density, and electromagnetic force gradually increased along the narrow gap, showing the maximum value at the welding point. In the temperature distribution in the narrow gap, the surface of the front end began to melt at an appropriate applied voltage, and the melting width was the largest at the welding point. As the applied voltage increased, the narrow gap became longer, and the appropriate gap length appeared in proportion to the applied voltage.

The high-frequency electric resistance welding (HF-ERW) process is widely used in the steel pipes production because it can weld at a high speed, has excellent weldability, and attains clean and precise shapes. However, for process improvement, analytic studies on electromagnetic field and temperature distributions, and selection of appropriate process variables are required. In this study, finite element analysis models that can analyze the electromagnetic field distribution and temperature distribution in the HF-ERW of a steel pipe were proposed, in consideration of the characteristics of the process, including electromagnetic phenomena localized to the workpiece surface and fast welding speed. By applying the proposed analysis models, changes in current density, magnetic flux density, generated heat density, and fused width in the pipe could be predicted according to changes in process variables such as the V angle of the strip, the electrode position, and the source voltage. Through comparison with the analysis and the limited-case experiment, the analysis result predicted the actual fused width fairly well, and the validity of the proposed model could be verified.

Heat treatment with carburized quenching process is widely used for automotive helical gear to improve its surface properties of hardness and strength. However, the gear can be deformed with the process over the allowable tolerance, which possibly makes noise, vibration and heat problems in operation. In this study, deformation of helical gear during heat treatment of carburized quenching was analyzed with a numerical method, incorporating coupled calculations of thermal conduction, carbon diffusion, phase transformation and thermal stresses. With the analysis, the effect of coolant temperature in quenching on the deformation was investigated. The result of the analysis revealed that the higher the coolant temperature became, the more change of helix angle and the more compressive stresses in the surface generated, because of delayed generation of martensite in the part.

The temperature distribution of the deposited droplet was predicted by using the finite element analysis and it was assumed that the droplet was axisymmetrical model. The analysis of residual stress was performed with the temperature data, which is obtained from the result. Axisymmetric droplet is deposited three times to consider the actual phenomenon of droplet deposition. The analysis of the temperature distribution is respectively performed whenever the axisymmetric droplet is laminated and the residual stresses of the laminated axisymmetric droplet are calculated with the value of the temperature distribution.