Fast thermal simulation of a heated crude oil pipeline with a BFC-Based POD reduced-order model

A proper orthogonal decomposition (POD) Galerkin reduced-order model (ROM) for unsteady-state heat conduction problems is proposed based on body-fitted coordinate (BFC), and the discrete format of spatial derivative of basis function which could enhance the calculation precision is employed in the present paper. Applying the POD reduced-order model into the fast thermal calculation of oil pipeline, such as the batch transportation of cold and hot oil as well as pipeline shutdown, the computing speed of POD reduced-order model is about 259 times and 177 times faster than that of traditional finite volume method (FVM) respectively. And from the calculation results, the reduced-order model is proved to enjoy good precision. Thus the reduced-order model holds great engineering application value and would be a very helpful tool to solve the time-consuming problem involved in the optimal design and risk assessment of oil pipeline.