Copper-magnesium alloy (CuMg0.4 alloy) was selected as the optimal material for pantograph-catenary systems, operating under extreme conditions including heavy rainfall, high winds, fog, and acid rain-induced corrosion. The synergistic effects of friction, wear, and corrosion significantly reduce service life, creating major maintenance challenges. This study investigates how ammonium salt concentration (Ac) in simulated acid rain affects tangential fretting wear and material degradation mechanisms in CuMg0.4 alloy. Experiments quantify frictional responses and wear characteristics under tangential fretting conditions. Machine learning (ML) and artificial intelligence (AI) models (Random Forest and DeepSeek) predict complex coupling effects, enabling data extrapolation beyond experimental parameters. This research establishes quantitative relationships between Ac and frictional failure mechanisms, providing a cost-effective predictive framework for material optimization.