The Soil Parameter Transformation Model Modification Based on Bayesian Method

Abstract In geotechnical engineering, empirical formulas and related relationships are site‐specific, and empirical formulas from different engineering sites cannot be universally applied to other sites. This paper proposes a Bayesian theory‐based correction method for soil parameter transformation models, constructs the likelihood function for the transformation model, and determines the prior distribution of the model parameters. This method can build a soil parameter transformation model with site specificity. The method proposed in this paper is applied to the correction of the soil unit weight transformation model, and the research results show that the method can effectively and accurately identify a soil parameter transformation model suitable for a certain geotechnical site. Moreover, by comparing the soil parameters estimated by the modified model with the true values, it demonstrates the effectiveness and correctness of this method.