This study investigates the thermo-mechanical behavior and residual stress characteristics of friction stir welding (FSW) in an aluminum inverter housing using finite element analysis (FEA). FSW experiments were first conducted under various tool rotation and traverse speed conditions, and temperature histories were measured using K-type thermocouples. The optimal process condition was identified through tensile testing, and the heat input was estimated by comparing experimental and numerical results. The estimated heat source was incorporated into a transient thermal elasto-plastic analysis to evaluate deformation and residual stresses in an inverter housing model. The results indicated that residual stress distributions varied depending on the welding start position. In particular, when welding started at P3 (near thick ribs and bosses) residual stresses were reduced by approximately 30% compared to P1, owing to the higher local stiffness and enhanced heat dissipation that mitigated temperature gradients. Conversely, welding initiated at P1, a flat region with insufficient reinforcement, resulted in higher stress concentrations. These findings confirm that the welding start position significantly influences residual stress behavior in inverter housings and provide fundamental insights for developing residual stress control strategies in FSW of large-scale components.

This study presents the results of compression, drop impact, and vibration durability analyses conducted to evaluate the mechanical reliability of Battery Pack Cases (BPCs) in electric vehicle (EV) systems. BPCs are essential structural components that must endure compressive loads, impact forces, and vibrational fatigue. Finite Element Analysis (FEA) was applied to a representative BPC model to assess deformation, impact resistance, and vibration endurance. The results indicate that the BPC maintained integrity within yield strength limits under compressive loading and effectively absorbed energy under drop impact. Furthermore, Power Spectral Density (PSD) analysis identified stress concentration regions, providing insights for structural optimization. Overall, the findings support the development of lightweight and reliable BPC designs for advanced EV applications.

Ear clamps are components used to securely fasten pipes and hoses in various industrial applications. They achieve clamping force by inducing plastic deformation at the ear region during installation, which can lead to accumulated structural damage and affect fatigue life. Moreover, the fatigue life is influenced by the design shape of the ear. Therefore, in this study, tensile and fatigue tests were conducted on two types of ear clamps with different ear geometries. Finite element analysis (FEA) was performed to obtain the stress distribution around the ear region, and these results were correlated with the experimentally obtained fatigue life. Based on this correlation, an S-N curve for simulation-based fatigue life estimation was established. This approach confirms the possibility of predicting the fatigue life of ear clamps with modified geometries using only finite element analysis, without the need for repeated fatigue testing.

Galaxy evolution studies require the measurement of the physical properties of galaxies at different redshifts. In this work, we build supervised machine learning models to predict the redshift and physical properties (gas-phase metallicity, stellar mass, and star formation rate) of star-forming galaxies from the broad-band and medium-band photometry covering optical to near-infrared wavelengths, and present an evaluation of the model performance. Using 55 magnitudes and colors as input features, the optimized model can predict the galaxy redshift with an accuracy of σ(Δz/1+z) = 0.008 for a redshift range of z < 0.4. The gas-phase metallicity [12 + log(O/H)], stellar mass [log(Mstar)], and star formation rate [log(SFR)] can be predicted with the accuracies of σNMAD = 0.081, 0.068, and 0.19 dex, respectively. When magnitude errors are included, the scatter in the predicted values increases, and the range of predicted values decreases, leading to biased predictions. Near-infrared magnitudes and colors (H, K, and H −K), along with optical colors in the blue wavelengths (m425–m450), are found to play important roles in the parameter prediction. Additionally, the number of input features is critical for ensuring good performance of the machine learning model. These results align with the underlying scaling relations between physical parameters for star-forming galaxies, demonstrating the potential of using medium-band surveys to study galaxy scaling relations with large sample of galaxies.