Rapid prediction and influence analysis of aero-engine rotor-stator clearance under maneuvering flight

The design level of the rotor-stator clearance (RSC) is a key factor affecting the aerodynamic efficiency and safety of the aero-engine. Excessive clearance causes gas leakage and efficiency loss, while insufficient clearance may cause rotor-stator rubbing and even affect flight safety. Therefore, it is urgent to analyze deformation under various maneuvering flights, identify key parameters affecting clearance, and provide data for clearance design. Based on the finite element model (FEM), the variation laws of the RSC under five typical maneuver loads are explored. A dataset covering 12 blade tip clearances and 9 labyrinth seal clearances under different loads is constructed, and then a lightweight BP neural network surrogate model is developed for rapid clearance prediction. The model efficiently predicts clearance variations with high accuracy under various operation conditions. Numerical results demonstrate that in the combined condition of constant-radius level turn superimposed with climbing, the climbing load plays a dominant role in the overall clearance. The surrogate model achieves second-level computation with maintained accuracy, offering key technical support for clearance design in turbofan engines.