US 12,450,757 B1
Method and system for obtaining fluid flow directions of tundish water model
Lifeng Zhang, Beijing (CN); Haojian Duan, Beijing (CN); Dinghan Li, Shenyang (CN); and Ying Ren, Beijing (CN)
Assigned to North China University of Technology, Beijing (CN); University of Science and Technology Beijing, Beijing (CN); and Northeastern University, Shenyang (CN)
Filed by North China University of Technology, Beijing (CN); University of Science and Technology Beijing, Beijing (CN); and Northeastern University, Shenyang (CN)
Filed on Jan. 3, 2025, as Appl. No. 19/009,062.
Claims priority of application No. 202410048945.4 (CN), filed on Jan. 12, 2024.
Int. Cl. G06T 7/254 (2017.01); G06T 7/246 (2017.01)
CPC G06T 7/254 (2017.01) [G06T 7/246 (2017.01); G06T 2207/10016 (2013.01); G06T 2207/10024 (2013.01); G06T 2207/20224 (2013.01)] 5 Claims
OG exemplary drawing
 
1. A method for obtaining fluid flow directions of a tundish water model, comprising following steps:
building the tundish water model with a preset similarity ratio with a tundish prototype;
carrying out an ink diffusion experiment in the tundish water model, and shooting a video of an ink diffusion process;
carrying out a binarization processing on the video of the ink diffusion process frame by frame to obtain binarized images;
wherein a method for obtaining the binarized images is as follows:
obtaining a frame at a moment before adding ink in the video of the ink diffusion process;
subtracting RGB values of each frame at a moment after adding the ink from RGB values of the frame at the moment before adding the ink to obtain RGB results;
obtaining images with background removed based on a range value of RGB and the RGB results;
re-obtaining ink distribution maps at different moments based on the images with the background removed; and
assigning values to different grayscale areas in the ink distribution maps at different moments based on a preset grayscale threshold to obtain the binarized images;
obtaining a time grayscale map based on grayscale values of the binarized images;
wherein a method for obtaining the time grayscale map is as follows:
concentrating the binarized images of the ink distribution maps at different moments on an image according to the grayscale values increasing with time to obtain the time grayscale map;
obtaining a velocity vector map of the tundish water model based on grayscale gradients of the time grayscale map; and
obtaining a moving direction of the ink based on the velocity vector map, and obtaining a flowing direction of a fluid in the tundish water model based on the moving direction;
wherein a method for obtaining the moving direction of the ink is as follows:
carrying out finite splitting on pixels of the time grayscale map to obtain a plurality of grids;
calculating a grayscale average value of each of the grids;
based on the grayscale average value, adopting a principle of minimum grayscale gradient, and obtaining a grid with a minimum grayscale value as a first grid within a preset grid range;
comparing grayscale gradients of a grid opposite to the first grid to obtain a grid with a smallest grayscale gradient with the first grid as a second grid; and
connecting an arrow between the first grid and the second grid to obtain the moving direction of the ink.