Inverse optimization design of heat transfer performance for the reactor pressure vessel insulation structure based on reduced-order model

BackgroundThermal leakage in reactor pressure vessel (RPV) insulation structure poses safety risks, and traditional trial-and-error parameter optimization methods are inefficient for handling complex variables.PurposeThis study aims to develop a proper orthogonal decomposition (POD)-based inverse design methodology to enhance the thermal performance of RPV insulation structure.MethodsFirstly, three design variables (air supply temperature, RPV outer wall temperature, and installation gap) and their corresponding level numbers were systematically determined, and three optimization objectives were defined as average heat flux (qave), average temperature (Tave) on the RPV insulation's outer wall, and maximum local temperature (Tmax) of the concrete. Then, a comprehensive dataset comprising 24 computational fluid dynamics (CFD) simulations was generated through full factorial design method, with these computational results employed as POD snapshots. Finally, the validated POD method was employed to reconstruct 275 working condition cases, and a dimensionless comprehensive score (FPOD) on a percentage scale was established to evaluate the overall heat transfer performance based on the three optimization objectives.ResultsCalculation results show the POD reconstruction achieved high fidelity is achieved by the POD reconstruction, with an average determination coefficient (R2) exceeding 0.95 and mean relative deviation (MRD) below 2.87%, while reducing CPU time is reduced by 99% compared to CFD. The optimal variable combination (2 mm-16 ℃-309.5 ℃) yieldeds a comprehensive score (FPOD) of 97.4, improving overall heat transfer performance by 59.2% over the original design.ConclusionsResults of this study demonstrate that the POD-based inverse design methodology provides an efficient and accurate approach for RPV insulation structure optimization. The optimal parameter combination achieves significant improvements in thermal performance, with qave reduced to 84.89 W·m-2, Tave to 35.21 ℃, and Tmax​to 72.86 ℃, which significantly enhances the safety and operational efficiency of RPV insulation structure.