RESUMEN
Direct laser deposition (DLD) requires high-energy input and causes poor stability and portability. To improve the deposited layer quality, conducting online measurements and feedback control of the dimensions, temperature, and other melt-pool parameters during deposition is essential. Currently, melt-pool dimension measurement is mainly based on machine vision methods, which can mostly detect only the deposition direction of a single melt pool, limiting their measurement range and applicability. We propose a binocular-vision-based online measurement method to detect the melt-pool width during DLD. The method uses a perspective transformation algorithm to align multicamera measurements into a single-coordinate system and a fuzzy entropy threshold segmentation algorithm to extract the melt-pool true contour. This effectively captures melt-pool width information in various deposition directions. A DLD measurement system was constructed, establishing an online model that maps the melt-pool width to the offline deposited layer width, validating the accuracy of the binocular vision system in measuring melt-pool width at different deposition angles. The method achieved high accuracy for melt-pool measurements within certain deposition angle ranges. Within the 30°-60° measurement range, the average error is 0.056 mm, with <3% error. The proposed method enhances the detectable range of melt-pool widths, improving cladding layers and parts.
RESUMEN
With the development of industrial automation, articulated robots have gradually replaced labor in the field of bolt installation. Although the installation efficiency has been improved, installation defects may still occur. Bolt installation defects can considerably affect the mechanical properties of structures and even lead to safety accidents. Therefore, in order to ensure the success rate of bolt assembly, an efficient and timely detection method of incorrect or missing assembly is needed. At present, the automatic detection of bolt installation defects mainly depends on a single type of sensor, which is prone to mis-inspection. Visual sensors can identify the incorrect or missing installation of bolts, but it cannot detect torque defects. Torque sensors can only be judged according to the torque and angel information, but cannot accurately identify the incorrect or missing installation of bolts. To solve this problem, a detection method of bolt installation defects based on multiple sensors is proposed. The trained YOLO (You Only Look Once) v3 network is used to judge the images collected by the visual sensor, and the recognition rate of visual detection is up to 99.75%, and the average confidence of the output is 0.947. The detection speed is 48 FPS, which meets the real-time requirement. At the same time, torque and angle sensors are used to judge the torque defects and whether bolts have slipped. Combined with the multi-sensor judgment results, this method can effectively identify defects such as missing bolts and sliding teeth. Finally, this paper carried out experiments to identify bolt installation defects such as incorrect, missing torque defects, and bolt slips. At this time, the traditional detection method based on a single type of sensor cannot be effectively identified, and the detection method based on multiple sensors can be accurately identified.
RESUMEN
Beryllium is widely used in the manufacturing of precision instruments because of its high thermal and mechanical properties. However, because beryllium is expensive, and processing it generally uses subtractive manufacturing methods, the cost is high, the utilization rate of cutting the materials is low, and the processing is difficult. Additionally, it is extremely prone to cracking, brittle fracturing, and fracturing during the machining process. In this paper, a new method for manufacturing beryllium laser additives under a pressure atmosphere is proposed. Via the single-point and single-pass laser melting of beryllium materials in an inert gas (Ar) pressure atmosphere, the results of the experiments conducted in the pressure range of 1 to 30 bar indicated the following: (1) beryllium can absorb the laser and form a molten pool, and the contour area of the upper surface of the molten pool is approximately 80% of that of 304 stainless steel under the same energy input; (2) severe oxidation occurs on and near the molten pool surface under low pressure, and oxidation is eliminated when the pressure is increased; (3) as ambient pressure increases, the surface profile of the molten pool gradually exhibits an irregular shape, and the cracks on the surface of beryllium change from "divergent" to "shrinkage", which can eliminate cracking. At higher pressures, the "small hole" phenomenon in the molten pool disappears, forming a wide and shallow molten pool shape that is more conducive to stable deposition. The experimental results indicate that the laser-additive manufacturing of beryllium in a pressure atmosphere is a meaningful developmental direction for beryllium processing in the future.
RESUMEN
Geometric characteristics provide an important means for characterization of the quality of direct laser deposition. Therefore, improving the accuracy of a prediction model is helpful for improving deposition efficiency and quality. The three main input variables are laser power, scanning speed, and powder-feeding rate, while the width and height of the melt track are used as outputs. By applying a multi-output support vector regression (M-SVR) model based on a radial basis function (RBF), a non-linear model for predicting the geometric features of the melt track is developed. An orthogonal experimental design is used to conduct the experiments, the results of which are chosen randomly as training and testing data sets. On the one hand, compared with single-output support vector regression (S-SVR) modeling, this method reduces the root mean square error of height prediction by 22%, with faster training speed and higher prediction accuracy. On the other hand, compared with a backpropagation (BP) neural network, the average absolute error in width is reduced by 5.5%, with smaller average absolute error and better generalization performance. Therefore, the established model can provide a reference to select direct laser deposition parameters precisely and can improve the deposition efficiency and quality.