Flow and heat transfer characteristics of twin-web disk with internal cross-ribs: a multi-objective optimization approach

With the increasing demand for higher thrust-to-weight ratios, raising the turbine inlet temperature has become the most direct and effective approach. The Twin-Web Disk (TWD) has garnered attention due to its light weight, enhanced rotational speed, and cooling performance, and also leading to various innovative structures. However, the complex temperature gradients and flow structures within high-speed rotating cavities remain insufficiently investigation. This study proposes a TWD with internal cross-ribs and investigates its intricate flow and heat transfer characteristics. The pressure distribution, temperature gradients, and wall heat transfer coefficients are systematically evaluated, incorporating dimensionless parameters such as the Nusselt number and the Omega vortex detection criterion. A design of experiments approach is employed to analyze the effects of cross-rib parameters (angle, thickness, height, and distribution density) and rotational conditions on these thermal-fluid characteristics. Based on these insights, a multi-objective optimization of the cross-rib structure is conducted using the proposed K-means Optimizer combined with Active Learning Kriging and the Non-dominated Sorting Genetic Algorithm combined with Active Learning Kriging to enhance pressure distribution and cooling performance. The results indicate that the optimized structure can reduce the peak temperature by up to 10.07 K. This research provides a theoretical foundation for optimizing aeroengine cooling systems, contributing to improved thermal performance and structural safety.