Simulation models can be effectively used to predict welding distortion and residual stress in laser welding systems. To this end, developing a sophisticated heat source model for finite element analysis (FEA) is crucial. This study proposes a novel multi-layered heat source model for FEA to predict the melting zone of the welds in fiber laser welding. The proposed model has multiple degrees of freedom sufficient to simulate various heat sources. A systematic approach by solving inverse problems with global optimization is devised to estimate heat source model parameters. To achieve this, a simplified FEA model is incorporated to accelerate the simulation process. Experimental results using 9% nickel steel indicate that the proposed framework provides more accurate predictions on the melting zone of the welds than existing methods while reducing the computational cost by 1/2000. The proposed framework can be used by field engineers to find the relevant heat source subjected to on-site welding conditions, minimizing trials and errors. With this method, fast and accurate welding simulations can be performed, ultimately leading to improved productivity.

Portable electronics makers have introduced liquid damage indicators (LDIs) into their products to detect warranty abuse caused by water damage. However, under certain conditions, these indicators can exhibit inconsistencies in detecting liquid damage. This study is motivated by the fact that the reliability of LDIs in portable electronics is suspected. In this paper, first, the scheme of life tests is devised for LDIs in conjunction with a robust color classification rule. Second, a degradation model is proposed by considering the two physical mechanisms—(1) phase change from vapor to water and (2) water transport in the porous paper—for LDIs. Finally, the degradation model is validated with additional tests using actual smartphone sets subjected to the thermal cycling of −15 °C to 25 °C and the relative humidity of 95%. By employing the innovative life testing scheme and the novel performance degradation model, it is expected that the performance of LDIs for a particular application can be assessed quickly and accurately.

Abstract In the fault diagnosis of rolling element bearings (REBs), spall size is a typical indicator of fault severity. Conventionally, spall size estimation relies on expert-knowledge-based or data-driven approaches. Expert-knowledge-based approaches require accurate assumptions about spall-excited events, making them challenging to apply in field environments. In contrast, data-driven approaches often struggle with insufficient training data and limited generalization across various operating conditions. To address this challenge, this paper proposes a frequency-enhanced neural network (FENN) with a hybrid spall-size estimator (HSSE). The proposed FENN employs both one-dimensional convolution in the time domain and Fourier convolution on the frequency magnitude, while preserving phase information in the frequency domain to enhance frequency components that are associated with spall in REBs. The novel HSSE proposed here integrates a data-driven spall-size estimator and an expert-knowledge-guided spall-size estimator to capture spall entry and exit events between rolling elements and raceways. Model validation results, which analyzed both simulation and experimental data from roller and ball bearings, demonstrate that the proposed approach provides accurate predictions of spall size, even with limited training data. Additionally, it is confirmed that the proposed model identifies the mechanical frequencies associated with spall events, providing interpretable results from raw vibration signals without requiring further processing.

This study presents a scale-free and phase-agnostic method for detecting interturn short-circuit faults (ISCFs) in the stator windings of permanent magnet synchronous motors (PMSMs). The proposed method integrates denoising and feature extraction with a crest-factor-based health indicator. A wavelet-based denoising method with average peak extraction is devised to enhance the sensitivity to fault-induced current deviations while suppressing external noise. The crest factor is modified to enable phase-agnostic detection across stator windings. To eliminate scale dependency associated with motor capacity, the modified crest factor is further calibrated using baseline data obtained under healthy conditions. The resulting health indicator, namely max-min crest factor, enables consistent and reliable ISCF detection without requiring motor-specific parameters or prior knowledge of the faulty phase. The effectiveness of the proposed method is validated through multiple case studies, including (1) numerical simulations, (2) laboratory-scale testbed experiments on small-capacity PMSMs (100 W and 1 kW), and (3) full-scale testing on an 82.5 kW PMSM under realistic operating conditions. The experimental results demonstrate that the method reliably detects ISCFs regardless of fault location or motor scale.

Respiratory viruses, such as influenza A virus and SARS-CoV-2, continue to pose significant global health challenges. Current antivirals, which are often specific to a single virus, face limitations due to rapid mutations and the emergence of new strains. In this study, we introduce styrene maleic acid copolymer lipid particle nanodiscs (SMALP-NDs) as a broad-spectrum antiviral platform that employs a dual mode of action. First, SMALP-NDs bind to positively charged viral proteins via their negatively charged surfaces, thereby blocking viral entry. Second, they induce the collapse of viral envelopes under acidic conditions similar to those in the endosome, leading to virus inactivation via a cell-mediated mechanism. SMALP-NDs demonstrated broad-spectrum antiviral activity against influenza A/B and multiple SARS-CoV-2 variants, including Omicron JN.1, as well as herpes simplex virus types 1 and 2 and vaccinia virus, underscoring their versatility. Intranasal administration of SMALP-NDs has successfully protected mice from lethal H1N1 and H5N2 influenza A viruses as well as SARS-CoV-2. These findings underscore that SMALP-NDs effectively counteract the increasing positive charge of emerging viral proteins through their negatively charged surfaces while leveraging pH-responsive virus inactivation mechanisms to achieve high antiviral efficacy with low toxicity, offering a significant advantage over traditional antiviral nanomaterials.