The fitting method of parameter distributions in Geotechnical engineering under small sample

The deficiencies existing in Geotechnical parameters distribution fitting using the classical probability distribution are analyzed and the stochastic weighting method is employed to improve the the small samples of Geotechnical parameters into big samples, thus sloving the problems caused by the small sample. The maximum entropy principle and the big samples improving from small sample are used to generate probability density functions of Geotechnical parameters. It is more rational and scientific with the method proposed in the paper through the Kolmogorov-Smirnov Test. It is demonstrated through the example that the method presented in this paper not only successfully escapes the reliance of the classical probability distribution functions, but also make the computation result more closed to the fact due to the data coming from the big sample improving from the small sample, which has important engineering significance.