Safety accidents related to falls and collisions involving strollers occur every year. To address this issue, an automatic brake system has been developed and tested for strollers. Previous systems were only functionally verified and did not confirm structural safety until the brakes were activated and came to a stop. Given that this system is a safety-critical device, a prototype was manufactured and tested to ensure the device's safety during brake operation. Additionally, structural analysis was conducted using the collected data to identify potential vulnerabilities.

In the rehabilitation of upper limb function impaired by stroke, facilitating the coordinated activation of multiple muscles is desirable. This study aims to analyze the coordination patterns of the tonic and phasic components of EMG during a reaching task and to investigate how the phasic component changes in relation to reaching speed. The analysis focused on the shoulder and elbow joints. EMG was recorded at five different speeds, with the slowest speed selected to represent the tonic component. The tonic component was then removed from the total EMG at the other four speeds to extract the phasic component. Correlation coefficients were calculated between the tonic component and joint angles, as well as between the phasic component and joint angular accelerations. For the tonic component, as joint angle increased during reaching, muscle activation also increased to counteract gravitational moments and enhance joint stiffness. For the phasic component, as reaching speed increased, the correlation between acceleration-deceleration patterns and muscle activation also increased. This suggests a greater synergistic contraction for enhanced acceleration and deceleration, as well as increased antagonistic contraction to ensure dynamic stability during faster movements

This study aims to optimize the process conditions for high-density polyethylene (HDPE) additive manufacturing through a systematic analysis of key variables, including material selection, layer height, feed rate, melting temperature, and bed temperature. By exercising precise control over these variables, optimal conditions were established, which included a melting temperature of 240oC, a welding speed of 150 cm/min, and a material throughput of 5.66 kg/h. Furthermore, the process was refined by implementing a zig-zag layering method, which significantly improved the stability, bonding strength, and overall mechanical properties of the final HDPE products. The effects of these optimized process conditions were assessed through a series of mechanical tests, such as tensile tests, impact tests, and heat deflection temperature (HDT) tests. As a result, the defined process conditions yielded excellent mechanical performance, achieving a tensile strength of 21.15 MPa, an impact strength of 320 J/m, and an HDT of 93oC. Overall, this study illustrates the enhancement of HDPE additive manufacturing quality through the optimization of process conditions. The strategic implementation of these optimized variables, along with advanced extrusion module design, demonstrates the potential for producing high-quality and cost-effective HDPE products, thereby underscoring their enhanced marketability and performance potential.

This study focuses on preventing folding defects in the forging process of parachute harness parts. Through three- dimensional finite element analysis, it was determined that folding defects arise from uneven metal flow and timing differences in the filling of various regions. To address these issues, a preform die was designed and evaluated using multi-stage forging simulations. The results indicated that the preform die facilitated uniform metal flow, preventing folding defects and ensuring consistent filling across all key areas. To verify the simulation results, surface and cross-sectional metal flow analyses were conducted. Additionally, the preform die reduced the maximum die load, which is expected to extend die lifespan and improve overall process efficiency. These findings demonstrate that precise control of metal flow and the application of a preform die can significantly enhance the quality and durability of forged components, providing valuable insights for improving forging processes across various industries

In this study, we propose a deep learning-based method for large-area inspection aimed at the high-speed detection of micro hole diameters. Micro holes are detected and stored in large images using YOLOv8, an object detection model. A super-resolution technique utilizing ESRGAN, an adversarial neural network, is applied to images of small micro holes, enhancing them to high resolution before measuring their diameters through image processing. When comparing the diameters measured after 8x super-resolution with the results from existing inspection equipment, the average error rate is remarkably low at 0.504%. The time taken to measure an image of one micro hole is 0.470 seconds, which is ten times faster than previous inspection methods. These results can significantly contribute to high-speed measurement and quality improvement through deep learning.

This paper addresses the issue of over-constrained assembly in mechanical designs using hole-pin patterns. When two hole-pin pairs are used, they can cause interference between components, leading to assembly failures. To mitigate this, designers often enlarge holes relative to pins to have a large float. However, when functional requirements do not permit significant float, field design engineers tend to add more assembly features, hoping them to mutually limit the float allowed by others. This numerical study employed two commercial tolerance analysis programs to demonstrate that these design changes could not sufficiently reduce float to justify added costs. Instead, this paper proposed an exactly-constrained design by replacing one of the holes with an elongated hole. Numerical analysis showed that this approach significantly reduced float compared to current design practices. This paper logically explains why this must be the case. It is hoped that this study contributes to the advancement of mechanical assembly design practices by adopting the exact constraint concep.

Glass Molding Process (GMP) is an effective method for producing precise optical elements such as lenses. This simulation study aimed to predict the distribution of temperature and stress within a lens during a multi-stage cooling process of GMP. To develop an accurate simulation model including molds and lens, thermal contact conductance and boundary conditions were determined by analyzing experimental and simulation results. The developed model was used to investigate changes in temperature and maximum principal stress within the lens, considering variations in cooling time, speed, and method at each cooling stage. Simulation results indicated that trends of maximum temperature difference and maximum principal stress within the lens were consistent over time. Results also showed that the maximum principal stress inside the lens increased significantly with additional cooling after uneven temperature distribution caused by a relatively short cooling time. Compared to simulation results of the cooling process involving contact only with bottom surface of the mold, contact cooling with both top and bottom surfaces showed decreased residual stress at the end of cooling and maximum temperature difference within the lens. However, the maximum principal stress could be higher during the cooling process involving both surfaces.

This paper proposed a CNC interpolator based on block overlap, capable of changing acceleration and deceleration time constants during continuous machining. The time constant can be set individually for each block through G-code commands. A velocity profile generation algorithm is proposed to set different time constants for both acceleration and deceleration phases. This algorithm can be applied to short blocks. The block overlap algorithm can be used for corner smoothing. A simulation model of the CNC interpolator was constructed to evaluate the proposed interpolation algorithm. Simulation results demonstrated that the proposed algorithm increased precision in areas with significant angular changes by adjusting time constants while simultaneously reducing machining time.

This paper presents a line-of-sight (LOS) stabilization control method for portable optical systems by analyzing fast steering mirror, image sensor, and gyro sensor system. To compensate for LOS errors caused by hand tremors in portable optical systems, we present the configuration of an image sensor-based LOS stabilization control system and a control strategy considering the phase delay effect caused by low sampling frequency of the image sensor. The phase delay effect of the image sensor caused restricted bandwidth, which limited the stabilization performance. To overcome such limitations, we present disturbance feedforward control using the gyro sensor and controller design method considering characteristics of the gyro sensor. Through overall system modeling, we constructed a control simulation model. The LOS stabilization performance against hand tremor disturbances was analyzed based on the proposed controller design. Simulation results demonstrated that integrating a gyro sensor-based disturbance feedforward control with the image sensor-based LOS stabilization control significantly enhanced the stabilization performance.

In this study, we developed a deep learning-based real-time fault diagnosis system to automate the weaving preparation process in textile manufacturing. By analyzing typical faults such as shaft eccentricity and rotational imbalance, we designed a data-driven fault diagnosis algorithm. We utilized tension data from both normal and faulty states to implement AI-based diagnostic models, including 1D CNN (Convolutional Neural Network), RNN (Recurrent Neural Network), and LSTM-AE (Long Short-Term Memory Autoencoder). These models enable real-time fault classification, followed by a comparative performance analysis. The LSTM-AE model achieved the best performance, with a classification accuracy of 99-100% for severe faults, such as 1.5 mm eccentricity and 100 or 150 g rotation imbalance, and 92.2% for minor faults like 1 mm eccentricity. This accuracy was optimized through threshold adjustments based on ROC curve analysis to select an optimal threshold. Building on these findings, we developed a GUI (Graphical User Interface) system capable of real- time fault diagnosis using TCP/IP (Transmission Control Protocol/Internet Protocol) communication between Python and LabVIEW. The results of this study are expected to accelerate the smartization of the weaving preparation process, contributing to improved textile quality and reduced defect rates, while also serving as a model for automation in other sectors.