Abstract
In practical engineering, the design data are uncertain. The data will deviate from the true value due to technical reasons such as measurement. It would result in the inaccuracy of crack fatigue life prediction. To deal with those problems, a regression model for crack fatigue life prediction is established in this study based on the conditional Bayesian theory. We use the prior distribution instead of posterior distribution in the iteration. The Monte Carlo sampling method is utilized to obtain the likelihood function and the prior distribution of the model. Then, the corresponding posterior distribution of the crack life can be obtained by likelihood and prior in iterative calculation. An engineering example of structural fatigue life of robotic is given to illustrate the application of proposed model. The effects of different parameter uncertainties on the posterior distribution of the model are compared. Also, the results in the least squares method are provided as reference.
Highlights
The machine learning based on probability theory has been applied in engineering widely, with the rapid development of artificial intelligence, pattern recognition, large data analysis and other theories [1]–[3]
We can describe the fatigue crack life model according to Eq (2) as: ae = f(X) + ε where outcomes ae have n observations varies, f(X) is the theoretical model of fatigue crack life, which is a function of parameters m, C and a0
Three issues have been discusseA novel time-variant reliability analysis method based on failure processes decomposition for dynamic uncertain structuresd in this paper
Summary
The machine learning based on probability theory has been applied in engineering widely, with the rapid development of artificial intelligence, pattern recognition, large data analysis and other theories [1]–[3]. The application of Bayesian method in machine learning can be divided into two main categories: the data classification and the data prediction [10], [15]–[17]. Chen and Wang proposed a weighted kernel density model based on original Bayesian model This method can select the optimal bandwidth and has a wide range of validity and applicability in the Bayesian classification [20]. Karandikar et al proposed a structural life prediction method based on Bayesian inference. Ma: Bayesian Inference Method and Its Application in Fatigue Crack Life Prediction including measurement methods, operating environment, sampling methods and so on [22]–[26].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
