A novel complex variable hyper-reduction model for efficient determination of convective heat transfer coefficient at the inlet of a steam turbine

To effectively determine the convective heat transfer coefficient of a steam turbine, the integration of the hyper- reduction model (HROM) with the complex variable derivative method (CVDM) is proposed for the first time to solve inverse heat conduction problem in this study. A hyper-reduction model is developed to quickly address the nonlinear transient heat conduction problem and is extended to the complex-variable domain to facilitate the implementation of the CVDM. The proposed complex-variable hyper-reduction model significantly reduces the computational complexity and enables the simultaneous acquisition of the simulated temperature and its gradient from the real and imaginary parts, respectively, within a single simulation. This study comprehensively validates the efficiency of the HROM, accuracy of the CVDM and robustness of the CVDM-based HROM. Results show that the HROM improves computational efficiency by nearly 20 times on the basis of finite element analysis, and the gradient of the simulated temperature on the convective heat transfer coefficient can be accurately solved from 10- 1 to10-9 step size. The proposed CVDM-based HROM exhibits good robustness and can accurately determine the convective heat transfer coefficient, even under a 5 % temperature fluctuation. It also shows excellent performance in identifying the dynamic convective heat transfer coefficient.