In this article, a new modeling and optimization procedure for Hansen solubility parameters (HSP) is developed. HSP values and the radius of interaction sphere for a solute can be determined by using this method based on experimental data at room temperature. The newly developed method could fit the experimental data better and get smaller radius of interaction sphere compared with the classical Hansen's method. The HSP values of carbon nanotubes and polymer matrix are calculated to show the accuracy of the proposed approach. The physical affinity between the carbon nanotubes and polymer in the filler-polymer matrix system composite has been evaluated by their HSP spheres interaction.