Abstract: As the core equipment of hydropower stations, the operational efficiency and health management level of hydroelectric generators are crucial to the overall performance of hydropower stations. Traditional fault diagnosis methods are inadequate in comprehensively reflecting the equipment status, suffering from slow diagnosis speed, low accuracy, and inaccurate fault localization, which fail to meet the demands of modern power station operation management. To address these issues, a novel fault diagnosis method for hydroelectric generators based on digital twin technology is proposed. This method constructs a digital twin model that includes a three-dimensional physical model, an operational mechanism model, and a data-driven model. Utilizing the Unity 3D engine, a fault diagnosis system is developed, enabling three-dimensional visualization of fault diagnosis results and maintenance procedures. This approach not only provides intuitive and rapid fault analysis and maintenance recommendations for station operators but also significantly enhances the operational efficiency and intelligent, digital management level of the station, demonstrating the immense potential of digital twin technology in the management of hydropower equipment.