Experimentally validated numerical model for multi-physics simulation of friction, wear, and noise in dry sliding pin-on-disc configurations

Tribology plays a crucial role in engineering, where friction, wear, and noise in sliding contacts impact efficiency and durability. This study develops a novel numerical framework for simulating dry sliding wear in a pin-on-disc setup using 6082 aluminium discs and 304 stainless steel pins. The model integrates Zhang-Meng-Chen multi-regime contact mechanics, Hurtado-Kim scale-dependent adhesion friction, data-driven asperity interlocking correction, Archard-based wear evolution, and symbolic regression-derived noise prediction, initialized with statistically equivalent rough surfaces from profilometry data. Validated against experiments at 10–20 N loads and 0.42–0.84 m/s speeds, the framework accurately predicts coefficient of friction (COF) transitions from adhesion- to interlocking-dominated regimes, contact area evolution, asperity counts, wear volumes, and cumulative sound pressures, with mean relative errors below 16%. Results reveal load-speed dependencies in friction mechanisms, surface topography changes, and acoustic emissions. This approach advances tribological modelling by linking microscopic interactions to macroscopic observables, paving the path for non-invasive machinery health monitoring through noise signals. Future enhancements could include thermal and debris effects. • Developed a novel multi-physics numerical framework integrating ZMC contact mechanics, HK scale-dependent adhesion friction, data-driven asperity interlocking correction, Archard wear evolution, and symbolic regression noise prediction for dry sliding pin-on-disc simulations. • Initialized with statistically equivalent rough surfaces from experimental profilometry data, accurately predicting COF transitions from adhesion- to interlocking-dominated regimes, contact area/asperity evolution, wear volumes, and cumulative sound pressures with mean relative errors below 16%. • Revealed load-speed dependencies: higher loads enhance interlocking friction while reducing adhesion contributions, leading to increased wear depths and acoustic emissions. • Demonstrated superior noise prediction over existing models, enabling correlation of noise signals with wear and COF for non-invasive machinery health monitoring. • Advanced tribological modeling by linking microscopic asperity interactions to macroscopic observables, with potential extensions to thermal and debris effects.