Objective With the rapid development of science and technology, there is increasing demand for high-precision workpieces in various fields, especially in space shuttles, aero-engines, space station, and medical fields. As one of the important branches of additive manufacturing, laser deposition-manufacturing technology plays an important role in high-precision and high-intensity manufacturing. In the laser deposition-manufacturing process, owing to the effects of some factors, such as heat accumulation, inevitably produces edge collapse, and the surface concave and convex inequality forming size deviation value phenomenon, resulting in a large deviation between the actual morphology and the ideal morphology. It affects the forming accuracy of the workpiece, and after multiple stacking manufacturing, the more concave the concave, the more convex the convex, which hinders the further progress of deposition manufacturing. Currently, research institutions and universities globally mainly focus on the optimization of forming processes, analyses of the structures and performance of the formed parts, and stress distribution during the forming process. There are only a few studies on improving the forming accuracy, such as morphology deviation and control. Meanwhile, it is vital to detect and control morphology deviations in sedimentary layers during forming processes.
Methods In this study, a high-speed profilometer was used to set up a sedimentary profile detection system, which was integrated into laser deposition-manufacturing equipment to detect and control sedimentary profile deviations. First, the high-speed profilometer was used to scan the surface of the sedimentary body, and the obtained three-dimensional morphology point cloud data were compared with the theoretical data of the sedimentary layer slices to extract the point cloud data that form the deviation area. Then, the deviated-area point cloud was layered and sliced, and the slices were converted to binary images by organizing the point cloud. The image boundary pixel points were extracted with the image boundary recognition algorithm and converted to coordinate points (i.e., the deviation contour point of the slice). The deviation contour feature line was fitted with the cubic B-spline curve. Finally, the accurate position of the deviated contour area in the original section contour area was determined, the filling space within the deviated contour area was changed, the forming track was filled, the deposition program was generated, and the deviated area was compensated. The flatness error on the surface of the sedimentary body before and after compensation was calculated, and the variation of the surface morphology deviation was analyzed.
Results and Discussions The results show that the morphology detection system can quickly obtain the morphology point cloud data of sedimentary bodies (Fig. 6). After the point cloud was denoised, a relatively ideal point cloud was obtained (Fig. 7). A pair of parallel planes was used to contain the denoised point cloud data to form the minimum containment area, and the flatness error value of the sample was obtained (Fig. 8). The morphologic point clouds are compared with the theoretical data of standard sediment slices to extract the deviated area point clouds (Fig. 9). The point cloud of the deviated region was layered and sliced, and the slices were converted into binary images. Then the deviated contour points of the slices were extracted and fitted using the image boundary recognition algorithm (Fig. 10). Finally, we propose a compensation path planning method based on changing the filling space of the deviated area to generate a compensation path, and the degree of depositional morphology deviation after compensation processing was significantly reduced compared with that before compensation (Fig. 11).
Conclusions Based on the above results, we draw the following conclusions. The laser deposition-manufacturing morphology detection system established can quickly scan the surface of the deposition to obtain the morphology point cloud data and the contour of the deviated region by processing the point cloud data. The accurate position of the deviated contour area in the original section contour area can be determined. Since the sedimentary shape is a sag deviation, the filling trajectory of the deviated contour area is filled and the compensation path is generated by reducing the filling space of the deviated contour area. The experimental results show that the deviation of the morphology of the sediment was compensated. The surface flatness error of the sample before and after compensation was 1.95 mm and 0.68 mm, respectively. This represents a 65.1% decrease in the fatness error. The degree of morphology deviation of the sample was significantly reduced, ensuring continuous deposition manufacturing and small machining allowance in the subsequent material reduction processes.
