Haibo Qiao, Yuling Lang, Decai Kong, Xiaoying Ma Chunhai Liu
CITIC Dicastal Co, Ltd, No 185, Long Hai Road, Economic and technological development zone,
Qin Huangdao, He Bei Province, China, 066011

EXTENDED ABSTRACT: As we all known, aluminum castings have significant differences in their internal microstructure due to the complex structure and non-uniform solidification conditions, which lead to different properties in different regions. At present, finite element analysis in the product design stage uses uniform material fatigue properties which resulting in simulation and test results were inconsistent. In this paper, an image recognition algorithm based on AI technology to automatically identify the microstructure (casting defects, secondary dendrite arm spacing, eutectic Si particles) of aluminum castings. Then, secondary dendrite arm spacing measurements and material fatigue tests were conducted on different areas of the counter pressure casting knuckles. a quantitative relationship between fatigue life, stress, and secondary dendrite arm spacing of counter pressure casting A356 alloy was established, and then the S-N curves corresponding to different SDAS is obtained. It is validated by experimental tests that the prediction results of the equation are in good agreement with the experimental results, and the maximum absolute value of the logarithm relative error of the fatigue life is below 2%. Finally, smooth and notched tensile and torsional specimens were extracted from different positions of the counter pressure casting knuckles for testing, and a material damage model based on stress triaxiality and fracture strain was established.
Keywords: Aluminum castings; secondary dendrite spacing; material fatigue model; stress triaxiality; material damage model