Micro-electrical discharge machining (micro-EDM) is a good candidate for processing micro-hole arrays, which are critical features of micro-electro-mechanical systems (MEMS), diesel injector nozzles, inkjet printheads and turbine blades, etc. In this study, the wire vibration of the wire electro-discharge grinding (WEDG) system has been analyzed theoretically, and, accordingly, an improved WEDG method was developed to fabricate micron-scale diameter and high-aspect-ratio microelectrodes for the in-process micro-EDM of hole array with hole diameter smaller than 20 μm. The improved method has a new feature of a positioning device to address the wire vibration problem, and thus to enhance microelectrodes fabrication precision. Using this method, 14 μm diameter microelectrodes with less than 0.4 μm deviation and an aspect ratio of 142, which is the largest aspect ratio ever reported in the literature, were successfully fabricated. These microelectrodes were then used to in-process micro-EDM of hole array in stainless steel. The effects of applied voltage, current and pulse frequency on hole dimensional accuracy and microelectrode wear were investigated. The optimal processing parameters were selected using response–surface experiments. To improve machining accuracy, an in-process touch-measurement compensation strategy was applied to reduce the cumulative compensation error of the micro-EDM process. Using such a system, micro-hole array (2 × 80) with average entrance diameter 18.91 μm and average exit diameter 17.65 μm were produced in 50 μm thickness stainless steel sheets, and standard deviations of hole entrance and exit sides of 0.44 and 0.38 μm, respectively, were achieved.

The present work introduces a laser-induced cavitation bubble technique for forming an axisymmetric structure (i.e., microgroove) and the dynamics of a cavitation bubble from initial expansion to the collapse stages that were also simulated. Furthermore, the shock wave signals and dynamic properties of the cavitation bubble were recorded using a hydrophone and a high-speed camera. The experiments on microgrooves formed by laser-induced cavitation bubble stamping were carried out, and the effects of laser energy, the initial position of the bubble, and the number of impacts on the microformability of aluminum sheets are discussed. The depth of the microgroove was investigated using experiments, and it was found that the process can serve as a rapid technique for impressing microfeatures on thin-sheet metals. The experimental results showed that as the initial position of the bubble increased, the deformation depth decreased. As the laser energy and number of impacts increased, the deformation depth increased. The results of the response surface experiments showed that a laser energy of 27 mJ, 3 impacts, and a bubble position of 3 mm were optimal for the process. By using the optimal parameters, flat and smooth microgrooves with a forming depth of 102.54 µm were successfully fabricated. Furthermore, the maximum thickness thinning of the microgroove section occurred at the entrance areas, and this area had the greatest hardness. This also indicated that the greatest amount of plastic deformation of the material and grain refinement occurred in this area. On the other hand, the aluminum foil did not undergo oxidation during the plastic deformation process. These results demonstrated that laser-induced bubble stamping is an advanced micromachining method with promising applications.

Grinding-assisted electrochemical discharge machining (G-ECDM) shows great potential for efficient and high-precision machining of insulating hard and brittle materials such as glass. However, research on ZrO ₂ ceramics, which have higher melting points and lower thermal conductivity, is still relatively lacking. To expand the application range of G-ECDM and address the thermal defects that occur in ZrO ₂ ceramics in conventional ECDM, this paper proposes a G-ECDM method based on constant spring force feedback feed. This approach aims to achieve faster removal of material affected by discharge heat through the grinding action of the abrasives, thereby avoiding excessive heat accumulation inside the material, and reducing the generation of thermally induced cracks with larger overcut. The feed force can be effectively controlled by means of a constant spring force feed, enabling better conditions for the synergistic action of grinding and discharging and preventing damage to the tool electrodes. In addition, previous research on G-ECDM has primarily focused on the grinding effect of abrasive tools, with few studies addressing the effects of the process on gas film formation and discharge. Therefore, this paper also investigates the impact of abrasive electrodes on gas film formation and discharge properties, while also examining the removal mechanism of ZrO ₂ ceramics during G-ECDM. Through a series of process parameter optimizations, high-quality small hole machining of ZrO ₂ ceramics was successfully achieved. This research provides an important basis for improving the theory of the G-ECDM process and expanding its application in high-temperature engineering ceramics.

Pre-wash sorting of medical textiles is essential for hospital infection control, yet accurate stain detection remains challenging because visually apparent stains often have blurred boundaries, whereas dried urine stains lack distinguishable optical features. This study proposes a medical textile stain detection method integrating chemically enhanced visualization with deep semantic segmentation. Dimethylaminocinnamaldehyde (DMACA) was used to convert latent urine stains into chemically developed stains with orange–red visual features. Based on the spatial color difference ΔE in the L*a*b* color space, 0.0183 mol/L was selected as the most suitable DMACA concentration among those tested. A dataset of 1974 images was constructed, including blood stains, chemically developed urine stains, medication stains, and uncontaminated textiles. A cascaded preprocessing strategy was applied to enhance stain boundaries and suppress textile texture noise, after which an Enhanced semantic segmentation model incorporating residual feature extraction, multiscale feature fusion, and transfer learning was used for pixel-level recognition. The IoU values for blood stains, chemically developed urine stains, and medication stains were 88.11%, 82.67%, and 89.62%, respectively. The average time required for image preprocessing and network inference was 15.39 ms per image. An input-level ablation comparison showed that DMACA-based color development increased the urine-stain IoU from 3.07% to 86.23%, demonstrating its substantial contribution to latent urine-stain detection. These results support the feasibility of integrating front-end chemical feature enhancement with back-end semantic segmentation for multiclass medical textile stain recognition under the current experimental conditions.