VHCF evaluation with BP neural network for centrifugal impeller material affected by internal inclusion and GBF region. (June 2022)

• Record Type: Journal Article
• Authors: Jinlong, Wang; Wenjie, Peng; Yongjie, Bao; Yuxing, Yang; Chen, Chen
• Is Part Of: Engineering failure analysis. Volume 136(2022)
• Publication Date: 2022-06
• Subjects: Very-high cycle fatigue -- GBF region -- BP neural network -- Fatigue life prediction; System failures (Engineering) -- Periodicals; Fracture mechanics -- Periodicals; Reliability (Engineering) -- Periodicals; Pannes -- Périodiques; Rupture, Mécanique de la -- Périodiques; Fiabilité -- Périodiques; Fracture mechanics; Reliability (Engineering); System failures (Engineering); Periodicals; Electronic journals; 620.112
• DOI: 10.1016/j.engfailanal.2022.106193
• Languages: English
• ISSNs: 1350-6307

Abstract

Highlights: The relationship between the GBF region size and the fatigue life is 'steep' under low stress amplitude states, while with the stress amplitude increases, this change trend is gradually flat. The neural network can be well applied to the prediction of VHCF life of centrifugal impeller for its strong fitting ability to discover and develop the connections behind complex data. The predicted results with BP neural network taking both internal inclusion size and GBF region size as the main variable are closer to the actual life. Abstract: The application of back propagation neural network (BP neural network) in very-high cycle fatigue life evaluation of centrifugal impeller to explore the effect of internal inclusion and granular bright facet (GBF) region on the fatigue life is a key and potential issue. Numerical simulation analysis of centrifugal impeller is carried out in this study to clear the mechanical state of centrifugal impeller in operation condition. Then, the designed very-high cycle fatigue test is conducted out; the test data and fracture morphology are analyzed to reveal the effect of internal inclusion and GBF region on the fatigue failure and life distribution. Then, with the comprehensive application of BP neural network, the fatigue life with different input parameters are predicted. In the case of different input parameters, the prediction changed and the very-high cycle fatigue life with the consideration of both internal inclusion and GBF region is very satisfactory. Study on neural network fatigue life prediction approach of centrifugal impeller in VHCF affected by internal inclusion and GBF region is novel for the further fatigue study in theoretical research and engineering practice for mechanical component and engineering metallic material.