Abstract：Chebyshe orthogonal polynomial approximation(CPA) method is introduced in this paper. Based on numerical approximation theory，CPA is employed to define the probabilistic density functions(PDF) of the geometry parameters of rockmass fissure network theory from the sample data. c2 test method is used to verify the availability of this method. Four classical distributions—the normal distribution，the log-normal distribution，the exponential distribution and Gama distribution，are compared with the corresponding CPA density functions generated from the moments. Little differences are found out between classical distribution and its corresponding CPA density function. The density distribution of joint tendency from field data of a freeway tunnel is taken as an example. The results of the case study show that the CPA density function is more accurate than fitting density function by using classical distributions.
