复杂换热结构的热特性分析和控制策略需要对温度场进行在线评估。本文针对冲击/气膜冷却壁,采用基于本征正交分解的降阶模型和稀疏传感器数据,重建系统的三维稳态温度场。本研究针对不同工况下的冲击/气膜冷却壁,建立了流固耦合模型,并完成有限体积法的数值模拟,生成高保真数据集。随后,利用本征正交分解提取三维温度场的空间分布特征,并通过提取正交基向量构建低维子空间。冲击/气膜冷却壁的温度分布被投影到低维子空间,实现模型降阶。通过建立传感数据与降阶模型之间的数学映射,实现复杂部件三维温度场的快速重构。通过对比重构结果、高保真仿真结果与实验结果,表明所提方法能够在毫秒级内快速、准确预测冲击/气膜冷却壁的温度场,相对误差在0.18%以内。这一方法为工程应用中温度场的实时监测和分析提供了可行的技术手段。

Accurate analysis and control of the thermal performance of complex heat exchange structures require real-time evaluation of their temperature fields. This study investigates impingement/effusion cooling walls and introduces a reduced-order model (ROM) based on Proper Orthogonal Decomposition (POD) and sparse sensor data to reconstruct three-dimensional steady-state temperature fields. A fluid-structure interaction model was developed for various operating conditions, and high-fidelity numerical simulations using the finite volume method (FVM) were conducted to generate a comprehensive high-resolution dataset. POD was then applied to extract the dominant spatial features of the temperature field, and orthogonal basis vectors were employed to construct a low-dimensional subspace. The temperature distribution of the impingement/effusion cooling walls was projected into this subspace, enabling efficient model order reduction. By establishing a mathematical relationship between the sensor data and the ROM, the complete three-dimensional temperature field of complex components was reconstructed rapidly. Comparative analyses with high-fidelity simulations and experimental data demonstrate that the proposed approach achieves millisecond-level predictions with a relative error of less than 0.18%. This method offers a practical and computationally efficient solution for real-time monitoring and analysis of temperature fields in engineering applications.