Abstract: In view of the existing problems of insufficient monitoring of equipment health status, lack of quantitative basis for operation and maintenance decisions, incomplete carbon emission evaluation system and others in the electromechanical system of Sima Mine, based on equipment health theory and life cycle evaluation method, a multi-source carbon emission metrology model for electromechanical equipment is constructed. A three in one low-carbon operation and maintenance strategy of "prediction-prevention-improvement" is proposed. By deploying 183 key equipment status monitoring networks, a fault prediction system based on BP neural network is established to achieve an early warning accuracy rate of 91.3%; A graded maintenance response mechanism is developed to improve the average health of equipment from 0.76 to 0.89; Energy-saving transformations such as variable frequency speed regulation, load optimization and others are implemented, the overall energy efficiency of the system is improved by 12.7%. A three-level carbon footprint evaluation system for equipment layer, system layer, and mine layer is constructed, and dynamic evaluation of carbon footprint is achieved by utilizing an improved fuzzy comprehensive evaluation method, the carbon emission intensity of the electromechanical system is reduced from 21.05 kg CO₂e/t to 18.36 kg CO₂e/t. The research results provide a scalable technical scheme for the low-carbon transformation of coal mine electromechanical systems.