冷热油交替输送热力快速计算研究 postprint

Abstract: A proper orthogonal decomposition (POD) Galerkin reduced order model based on body-fitted coordinate is applied to the fast calculation of the thermal process during cold and hot oil alternating transportation, and the implementation process is given in detail. Considering the uneven distributed heat conductivity coefficient in the thermodynamically involved region of the pipeline (the ratio of heat conductivity coefficients is as large as 300 around), a discretization scheme better fitting the physical meaning of the derivative of the basic function space is employed, instead of the traditional used central difference scheme. Trough all the cases devised, the averaged computation error by POD reduced order model is below 0.35℃, while the speed can reach 260 times of the traditional finite volume method.