Mine-hoisting wire ropes are critical load-bearing components that operate under repeated tensile loading, bending, contact, and environmental exposure. During service, surface degradation mechanisms such as wear, abrasion, corrosion, fretting, and lubrication-related deterioration can progressively reduce the effective metallic area and alter the mechanical properties of the rope. These changes may influence local stiffness, mass distribution, damping, and dynamic response, potentially providing measurable signatures for condition monitoring.
This study proposes a physics-informed neural network (PINN) framework for monitoring surface-induced degradation in mine-hoisting wire ropes through vibration measurements. An equivalent dynamic model is formulated to describe the transverse vibration of the rope using effective mass per unit length, tension, damping, and spatially varying mechanical properties. Surface degradation is represented through a localized reduction in effective metallic area and/or stiffness, providing a tractable connection between surface damage and the rope's dynamic behavior.
An inverse PINN is then developed to combine the measured vibration response with the governing rope dynamics and estimate degradation-related parameters, including fault location and degradation severity. The framework is evaluated under different sensor densities, noise levels, and operating conditions and is compared with conventional vibration-based analysis and data-driven machine-learning approaches. Particular attention is given to parameter identifiability, robustness, physical interpretability, and the limitations associated with representing complex wire-rope degradation using equivalent mechanical properties.
The study aims to establish a physics-informed connection between surface degradation mechanisms and measurable structural dynamics, providing a potential framework for non-destructive condition monitoring of mine-hoisting wire ropes.

Mine hoisting,wire ropes,Surface Degradation,Physics-Informed Neural Networks (PINNs),Condition Monitoring