Abstract: This paper focuses on the health monitoring of the liquid distribution pump in in-situ leaching uranium mines. An application method of fault prediction and equipment health intelligent maintenance cloud （iPHM） based on digital twin is proposed to improve the health management level of liquid distribution pump by means of diagnosis and early warning. Through data interconnection, the integration of physical entities, digital twin models, digital twin data and services is realized, and real-time monitoring, fault prediction and health management of the pump body are completed. The vibration in the twin model data analysis and the bearing fault feature extraction analysis of the envelope slice spectrum is analyzed to verify the effectiveness and superiority of the envelope slice spectrum in the bearing fault feature extraction. It provides a theoretical basis for the digital twin model analysis of the pump. In addition, the application service requirements of digital twin as an enabling technology of virtual and real integration in simulation, control, prediction, optimization and other aspects, as well as the connotation and key technology exploration of its connection and interaction are also explored. The purpose of this study is to improve the overall operation efficiency, maintenance economy and safety of the liquid distribution pump in in-situ leaching uranium mines, and to provide a useful reference for the practical application of digital twin technology in the mining field.