高岭土-砂土混合介质电导率预测模型适用性及误差评估OA北大核心CSTPCD

In order to improve the inversion accuracy of the conductivity prediction model for the structural properties of rock and soil,the conductivity of kaolin-sand mixed media with different clay mass fractions was measured under different pore fluids.According to the way of parameter acquisition,the existing conductivity prediction models are divided into fitting models and empirical models.The conductivity prediction values of each model are obtained by fitting calculation,and the error evaluation system is established by comparing with the measured values.The research results show that for kaolin-sand mixed medium,the prediction effect of the fitting model is better than that of the empirical model.In the empirical model,the predicted values of the generalized Archie model under different clay mass fractions are greater than the measured values.Therefore,when using the generalized Archie model for conductivity prediction,it is necessary to correct and adjust.The Waxman-Smits model is suitable for predicting the conductivity of mixed media with low clay mass fraction.The modified Archie model has a good prediction result for the conductivity of the mixed medium with medium clay mass fraction.The Rhoades model has the best prediction effect on the conductivity of the mixed medium under high clay mass fraction.