Statistics investigation of mechanical parameter of rock based on Bayesian reliability theory

Abstract

In order to improve forecast accuracy of bearing capacity of rock, the forecast method of bearing capacity of rock was constructed on Jeffreys prior using MCMC method of Bayesian theory. The proposed approach was used to estimate the parameters of Normal distribution. Numerical simulation study was used to produce the pseudo samples. The maximum likelihood parameter estimation method and Bayesian statistical theory were used to estimate the forecast value of the Normal distribution, which has been done by comparing with the theoretical value of the pseudo sample of Normal distribution. The result indicates that the forecast model of Normal distribution is optimal than the maximum likelihood estimation, and a significant improvement were noticed with the number increasing of pseudo sample. At last, the proposed method was applied to estimate the Normal distribution of bearing capacity of rock, which shows that the proposed method in this paper has higher precision and good applicability.