To address the challenges in laser-cladding molten pool image segmentation, such as inadequate multilevel feature extraction, poor multiscale spatial feature fusion, ineffective noise suppression, and limited segmentation adaptability, a laser-cladding-oriented molten pool segmentation framework based on spatial feature fusion and an attention mechanism (RAE-U-Net) is proposed. The framework integrates the U-Net architecture with the ResNet-50 residual network, adaptive spatial feature fusion, and efficient channel attention mechanism to enhance feature extraction, multiscale feature fusion, and channelwise feature representation for laser-cladding molten pool images. The proposed framework enables multilevel and multiscale feature representation and achieves high-precision molten pool segmentation within the investigated dataset. Experimental results show that RAE-U-Net achieves a mean intersection over union of 97.33%, with mean pixel accuracy, precision, and recall values of 98.82%, 98.41%, and 98.82%, respectively. These results demonstrate the effectiveness of the integrated framework for laser-cladding molten pool image segmentation and provide a vision-based solution for molten pool segmentation under the investigated process conditions.
Low-pressure die-casting (LPDC) aluminum (Al) alloys have been widely utilized in emerging sectors such as new energy vehicles, owing to their benefits in integrated casting and lightweight design. However, the surface quality of LPDC Al alloys often fails to meet direct application requirements. This study applies laser polishing to enhance the surface quality of LPDC A356 alloy. A continuous-wave fiber laser (1060 nm) was used at a fixed scanning speed of 1000 mm/s, with laser power varied from 260 to 340 W to examine its effect on the surface quality. Results indicate that laser polishing significantly reduces surface roughness of the LPDC Al alloy from an initial 6.44 ± 1.70 down to 0.65 ± 0.06 μm, while small deep holes on the as-cast surface are no longer observed after polishing. Microstructural analysis reveals that the surface layer transitions from a structure consisting of α-Al and eutectic Si to α-Al and dispersed Si phases within the grains. The microhardness of the laser-polished layer increases by 89% compared to the substrate, which is consistent with the refined grain structure resulting from the high cooling rate. However, attention is required regarding the corrosion performance of laser-polished surface as it could be influenced by microstructural features such as the nonuniform distribution of fine Si particles in the remelting layer.
With the growing demand for high cleanliness and precision in metal surface treatment in modern industry, traditional paint removal methods have proven inefficient, environmentally unsustainable, and prone to substrate damage, failing to meet contemporary industrial standards. In contrast, laser cleaning technology has emerged as a promising alternative due to its superior efficiency, precision, and environmental compatibility. This study proposes a machine learning-based framework for predicting the quality of laser-induced paint ablation. An experimental platform employing an infrared nanosecond laser was developed to collect data on paint removal performance under varying process parameters, forming the basis for the machine learning dataset. By integrating support vector regression, artificial neural networks, and random forest (RF) algorithms, nonlinear mapping models were established to correlate process parameters with cleaning quality metrics, including removal rate, surface roughness, and paint removal thickness. Bayesian optimization was utilized for parameter tuning. Experimental results demonstrate that the single-objective RF model achieves exceptional predictive accuracy, with a test set R2 value of 0.9465 (single-objective model). Building on these findings, an intelligent optimization framework combining RF with Bayesian optimization was developed, enabling the identification of high-performance parameter combinations. The optimized predictions yielded a removal rate of approximately 98.9%, a paint removal thickness of around 100.8 μm, and a surface roughness of about 2.74 μm. This work integrates multialgorithm modeling with intelligent optimization strategies, providing a robust theoretical and technical foundation for the intelligent control of laser cleaning processes.
This study examined through-thickness cutting of 110-mm-thick SUS304–SS400 clad steel (10 mm SUS304/100 mm SS400) using a 6-kW laser oxy-fuel and oxy-flame hybrid (LOXAFH) process. With SUS304 facing the cutting head, oxy-fuel heating formed no measurable entrance kerf. Three laser-assisted oxygen cutting (LASOX) trials failed to penetrate at 6 kW, 0.70 MPa(g) cutting-oxygen pressure, 50 mm min−1, and 10 mm standoff distance. At the same principal settings, one LOXAFH trial achieved full penetration with an O2–liquefied petroleum gas flame and approximately 30 s of stationary preheating. Two additional LOXAFH trials produced through cuts at selected settings. Across these three cuts, top and bottom kerf widths were 6.80–7.75 and 9.55–10.38 mm, respectively. At otherwise nominally identical LOXAFH settings, one 6-kW trial penetrated, whereas eight 3-kW trials failed. At 3 kW, three SS400-up trials penetrated, while eight SUS304-up trials failed. These comparisons indicate that laser power and incident material affected penetration. The matched LASOX–LOXAFH comparison supports feasibility, but the effects of stationary preheating and continuous flame assistance could not be separated. Improved oxygen access remains a possible explanation because in-kerf flow and cut-front motion were not measured. Repeatable penetration was observed at selected conditions, but repeatability at a fixed successful setting for the standard SUS304-up plate remains unverified. Comprehensive cut-quality assessment and suitability for nuclear dismantling require further evaluation.
Traditional water-assisted Er:YAG laser ablation of hard bone tissue is limited by the difficulty of controlling the water-layer thickness. Because water strongly absorbs 2.94 μm Er:YAG laser radiation, an excessively thick or stagnant water layer attenuates the incident laser energy and can reduce ablation efficiency. In this study, an off-axis water-air-jet-assisted Er:YAG laser ablation method is proposed to regulate the water layer on the surface of hard bone tissue. The air jet impinges on the water layer in the ablation region, and the water-layer characteristics are modulated by water pressure, air pressure, and air-nozzle diameter. A computational fluid dynamics model was established to simulate air-jet modulation of the water layer and to analyze the resulting water-layer thickness, water volume fraction, and flow velocity. Comparative ablation experiments were then performed under air-jet, water-jet, and off-axis water-air-jet assistance. Relative to air-jet assistance, the off-axis water-air jet markedly reduced visible carbonization and the optically observed heat-affected region. Relative to water-jet assistance, the ablation depth and hole diameter increased by 5.8% and 13.5%, respectively. Single-factor experiments further showed that water pressure, air pressure, and air-nozzle diameter affected the ablation morphology and depth in a nonlinear manner through changes in water-layer thickness and water volume fraction. These findings demonstrate that off-axis water-air-jet modulation can balance cooling and laser-energy transmission during Er:YAG laser ablation of hard bone tissue. The assessment of thermal effects was limited to optically visible carbonization and heat-affected morphology; histological evaluation will be required to quantify microscopic thermal injury.
