Search

Laser vision seam tracking enhances robotic welding by enabling external information acquisition, thus improving the overall intelligence of the welding process. However, camera images captured during welding often suffer from distortion due to strong noises, including arcs, splashes, and smoke, which adversely affect the accuracy and robustness of feature point detection. To mitigate these issues, we propose a feature point extraction algorithm tailored for weld images, utilizing an improved Deeplabv3+ semantic segmentation network combined with EfficientDet. By replacing Deeplabv3+'s backbone with MobileNetV2, we enhance prediction efficiency. The DenseASPP structure and attention mechanism are implemented to focus on laser stripe edge extraction, resulting in cleaner laser stripe images and minimizing noise interference. Subsequently, EfficientDet extracts feature point positions from these cleaned images. Experimental results demonstrate that, across four typical weld types, the average feature point extraction error is maintained below 1 pixel, with over 99% of errors falling below 3 pixels, indicating both high detection accuracy and reliability. [ABSTRACT FROM AUTHOR]

Hydraulic excavators, characterized by their high efficiency and versatility, are crucial equipment in engineering construction. During the automatic excavation process, the excavator may experience resonance phenomena, which can affect the stability of the entire system. In order to study the vibration harmonic response of the excavator arm in the process of reciprocating motion, firstly, the model of the excavator arm is designed in three‐dimensional software, and the finite element meshing software is used to perform hexahedral meshing of the excavator arm model and define the material properties. Secondly, the modal analysis of the excavator arm is carried out by finite element simulation software, and the displacement and acceleration harmonic response analysis is carried out using the modal superposition method. Finally, the vibration experiment of reciprocating motion with different frequencies is carried out on the experimental platform of the excavator arm. The simulation and experimental results show that the maximum value of the displacement and acceleration response of the excavator arm corresponds to the frequency near the second‐order intrinsic frequency, and the excavator arm is greatly affected by the second‐order intrinsic frequency; controlling the frequency of the reciprocating motion of the hydraulic cylinder within 13 Hz can effectively avoid the resonance phenomenon of the excavator arm device. [ABSTRACT FROM AUTHOR]

Facing severe aerodynamic and inner equipments' heat environment, the integrated thermal management platform composed of thermal protection system and thermal management system decides the closed performance of aircraft overall scheme, so the preliminary design scheme of integrated thermal management system which determines the feasibility of aircraft design needs to be rapidly simulated to judge its feasibility in the aircraft design stage. Taking electronic equipment cabin as an example, the fast simulation platform using modular design and one-dimensional simplification method is built by modularly packaging of heat source, heat transfer, thermal control and other simulation components, and is verified with simulation instance. The results show that the platform can quickly realize the model establishment, simulation and evaluation of aircraft electronic equipment cabin integrated thermal management scheme.

Objective: To compare the surface roughness of nanofilled dental composite resin and microhybrid composite resins after curing and polishing., Methods: A nanofilled composite (Z350) and 4 microhybrid composites (P60, Z250, Spectrum, and AP-X) were fabricated from the lateral to the medial layers to prepare 8 mm×8 mm×5 mm cubical specimens. The 4 lateral surfaces of each specimens were polished with abrasive disks (Super-Snap). Profilometer was used to test the mean surface roughness (Ra) after polishing., Results: P60 had the lowest Ra (0.125∓0.030 µm) followed by Z250 and Spectrum. The Ra of Z350 (0.205∓0.052 µm) was greater than that of the other 3 resins, and AP-X had the roughest surfaces. Under scanning electron microscope, the polished faces of P60 resin were characterized by minor, evenly distributed particles with fewer scratches; the polished faces of Z350 presented with scratches where defects of the filling material could be seen., Conclusion: The nanofilled composite Z350 has smooth surface after polishing by abrasive disks, but its smoothness remains inferior to that of other micro-hybrid composite resins.