Search

Aiming at the lack of technology for generator incipient condition monitoring and fault prediction, proposed SVM(Support Vector Machine) is introduced to generator incipient fault prediction. This paper takes the parametric faults of synchronous generator as an example, selects the output voltage as a monitoring signal, and combines with SVM regression prediction algorithms to achieve synchronous generators incipient fault prediction.

Fault tree modeling and analysis methods are being developed in academic and industrial. This paper designed a real-time diagnose method based on the .NET platform, combining the fault tree analysis (FTA) and real-time systems. Use SQL Server as the knowledge-base to store the information. Take the fault tree information and monitoring data as inputs, diagnose the real-time alarm of the system using minimal cut set algorithm (MCS).Finally test and verify the method by simulation.