Comprehensive multi-parameter sensitivity analysis of a semi-empirical model for airfoil self-noise prediction

Urban noise pollution driven by rapid infrastructural development and tightening environmental regulations has made aeroacoustic emissions from mechanical ventilation systems a critical concern for public health and environmental quality. Fan noise, encompassing both tonal and broadband components, significantly impacts acoustic comfort, necessitating a fundamental understanding of the aeroacoustic principles behind airfoil self-noise, which dominates broadband emissions from the blades. This study presents a comprehensive statistical sensitivity analysis of the parameters governing airfoil self-noise across an extended operational envelope, aiming to provide a structured basis for dimensionality reduction in aeroacoustic optimization loops. The semi-empirical prediction framework is validated against experimental literature data, and a two-stage Morris screening method is implemented to systematically quantify and rank the influence of multiple aerodynamic, geometric, and turbulence-related parameters on sound pressure levels. The first-stage screening establishes the absolute dominance of the Mach number, followed by chord length and trailing-edge bluntness thickness as the primary drivers of overall acoustic emissions. The second-stage analysis identifies the boundary layer displacement and thicknesses as critically important parameters directly governing turbulent boundary layer trailing-edge noise and laminar vortex shedding mechanisms. Furthermore, the Morris method successfully captures and quantifies the cross-coupling interaction effects among the input parameters, providing rigorous guidance for prioritizing key variables in advanced aeroacoustic modeling and boundary layer management.