The Fault Diagnosis of Blower Ventilator Based-on Multi-class Support Vector Machines

Blower ventilators are one of the main rotating equipment in the thermal power plant, supervising and forecasting the faults of the operating ventilator can significantly improve the safety and economy of ventilator as well as guarantee the normal operation of blower. In this paper, the basic principle of faults diagnosis and advantages of DAGSVM are analyzed, the knowledge library of ventilator operating faults is established and trained based-on DAGSVM. Taking a large-scale boiler blower as an example, the DAGSVM model is used to diagnose the actual operating faults, the result shown that DAGSVM can diagnose the common faults of ventilator effectively. Forecasting the ventilator operating circumstance by this method can improve the safety and economy of ventilator operating.