A comparative study of vision detection and numerical simulation for laser cladding of nickel-based alloy

The vision detection system was built to detect the molten pool of laser cladding(LC), then the captured images by detection system were processed with image-processing technology, finally the shape and the characteristic parameters were obtained in real-time. Nevertheless, the simulation analysis is a useful mean to predict the temperature contour in practice, so the three dimensional numerical model with moving heat source was built by APDL technique to obtain the temperature distribution and then extract the shapes and sizes of molten pool from the temperature contour by a threshold value of material's melting point. After experiments, the microstructure was examined to analyze the height of cladding track and height of HAZ, and dilution as well. By comparing the result of vision detection and simulation results, the section of cladding track is crescent shape, and results by vision detection agree fairly well with that of simulation. Meanwhile, the relationship between dilutions, area, width, length of molten pool and input power, scanning speed and pre-placed powder thickness were studied respectively. The results shown that the dilutions, area, width, length of molten pool will decrease with the increased scanning speed, while the increased power input will cause the increase of molten pool area, size and dilution. Thicker powder result in larger area, width and length of molten pool while less dilution.