Influence and Quantitative Analysis of Different Parameters on Adaptive Cycle Fan Performance

This study investigates the adaptive cycle fan (ACF), a key component enabling variable-cycle functionality in next-generation adaptive cycle engines (ACE). Despite its critical importance for whole-engine matching, the full operating-range performance of the ACF and the coupling effects among its parameters have not been systematically examined. This work addresses this gap. Owing to its dual-flowpath architecture and multiple adjustable variables, advanced modeling approaches are required; therefore, a neural-network-based surrogate model is developed to map parameter variations to ACF performance. Based on this model, the full operating-range performance of the ACF is analyzed. The constant-speed performance forms multi-line surfaces with distinct trends across rotational speeds. Core throttling provides wide-range total pressure ratio regulation, while bypass throttling enables broad bypass ratio modulation with relatively stable pressure ratio and efficiency. To interpret the neural network, the SHAP method is employed to quantify parameter sensitivity and multi-parameter coupling effects. Bypass outlet backpressure, core outlet backpressure, and front-fan tip clearance are identified as dominant factors, exhibiting strong coupling effects that must be jointly considered for optimal engine regulation. This study presents the first three-parameter coupling analysis of an ACF and provides guidance on ACE control design, component matching, and adjustable structure design.