Abstract: Abandoned coal mining areas in China are widely distributed across coal-bearing strata rich in groundwater resources. The diversity of mine types, variable hydrogeological conditions, complex transport pathways, and intricate hydrobiogeochemical evolution processes have resulted in poorly understood distribution patterns of acid mine drainage (AMD) and ineffective pollution control. AMD mitigation remains a central challenge in the ecological restoration of abandoned mining areas, facing four critical issues: (1) Mining disturbances cause complex evolution of aquifer media and dynamic fields, where pyrite oxidation and metal ion migration are governed by coupled geological-hydrobiological-microbial factors, posing challenges in mechanistic understanding and accurate modeling; (2) The efficiency of existing technologies for in-situ reduction of high-concentration characteristic pollutants (e.g., sulfates, Fe/Mn) is constrained by dynamic water quality and low-temperature conditions, significantly limiting their stability and long-term effectiveness; (3) Control systems lack dynamic intelligent regulation capabilities, with fragmented data perception-decision-execution loops failing to adapt to spatiotemporal heterogeneity in water quality and quantity; (4) Insufficient synergy between ecological restoration and pollution control hinders holistic recovery of mining ecosystem functions. Future innovations should focus on interdisciplinary integration to establish a "theoretical models-technological tools-engineering paradigms" innovation chain, advancing AMD control from empirical management to scientific regulation. This will provide Chinese solutions for global mining area restoration, specifically: (1) Revealing biogeochemical micro-mechanisms of AMD generation and developing coupled multi-field models for pyrite oxidation kinetics and metal migration; (2) Developing key enzyme inhibitors or biomimetic oxygen-barrier materials for targeted in-situ blocking of pyrite oxidation; (3) constructing a "digital twin + smart sensing" precision control platform for pollution evolution simulation and dynamic remediation optimization through multi-source data fusion; (4) Establishing a "water-soil-biota" synergistic regulation theory based on ecohydrological processes to elucidate feedback mechanisms between natural restoration and engineering interventions.