Currently, the measurement and prediction of the electrical transport performance of thermoelectric double-layer membrane is often based on the theory of parallel model. However, the conditions under which the parallel model can be used lack theoretical and experimental support and validation. In this work, the Seebeck coefficients of Cu/Si and Ag/Si bilayers under applied temperature difference are obtained by using finite element theory simulations with the help of COMSOL Multiphysics software and compared with the results from the parallel model. Whether the ends of the bilayer plated with a metal Pt layer, and the insertion of a high-resistance/low-resistance/insulation interface between the bilayers affect the Seebeck coefficient measurements of the bilayer are investigated. It is found that when there is no Pt at the hot end or cold end, the potentials on the Si side and Cu side at the high-resistance and electrically insulating interfaces are uniformly distributed along the direction of the temperature gradient, respectively, and the measured Seebeck coefficients are the same as the values of the materials’ own, respectively, and the thermal potential on the Cu side at the low-resistance interface varies uniformly with the probe spacing L, while the thermal potential on the Si side shows a non-uniform variation. With Pt, the thermal potentials on the Cu side and Si side are uniformly distributed along the direction of the temperature gradient, and the measured values on both Si side and Cu side are the same as the Cu Seebeck coefficients, regardless of the insulating/high-resistance/low-resistance interface. The Si/Ag and Bi/Ag bilayers are investigated experimentally. In the absence of Pt, the absolute value of the Seebeck coefficient on the Si side of Si/Ag bilayer decreases with temperature decreasing, but the absolute value of the Seebeck coefficient on the Ag side increases with temperature decreasing. In the presence of Pt, the Seebeck coefficients on both sides of the Bi/Ag bilayer membrane are equal.