Research on Condition Monitoring and Fault Early Warning Method of Turbine Generator Slip Ring Carbon Brush Based on Multimodal Data Fusion

The carbon brush system of the steam turbine generator slip ring is a key component of the excitation circuit, and its operating condition directly affects the safety and stability of the unit. In response to the problems of the traditional monitoring methods being single and the fault warning being lagging, this study proposes a state monitoring and fault warning method based on multi-modal data fusion. By integrating multi-source information such as infrared temperature measurement, video image analysis, vibration monitoring, and circuit models, a comprehensive performance evaluation system for the slip ring carbon brush system is constructed. A temperature-current-wearness coupling analysis model was established, and a multi-modal feature fusion algorithm based on deep learning was developed, enabling the early identification of typical faults such as abnormal carbon brush heating and poor contact. Experimental results show that this method can advance the fault warning time by more than 60%, with a diagnostic accuracy rate of 98.2%, significantly improving the equipment reliability. The research results provide a new technical means for the condition-based maintenance of power generation units and have significant engineering application value.