Regression analysis estimation of thermal conductivity using guarded-hot-plate apparatus

In order to utilize the guarded-hot-plate (GHP) apparatus to study the influence of heating direction and the significance of convection within pores of porous materials, the applicability of the GHP apparatus to estimate thermal conductivity of two specimens with different thermal conductivity was investigated. To predict the value of thermal conductivity of two different specimens, a new testing method is proposed using multiple linear regression analysis. In the new testing method, the issue of multicollinearity (statistical dependency between two predictors) was addressed by manipulating the top and bottom auxiliaries’ temperature. This proposed method requires that the temperature of top and bottom auxiliaries shall be different. Experimental results show that the proposed method can be used to measure accurately two specimens having same or different thermal conductivity.