Research On Monitoring And Fault Diagnosis Technology Of Uhp Eaf Dust-removing Air Blower

Air blower is a kind of machine, which can make use of mechanical energy to transport gas. It is an essential equipment steel mill and power plant. In practical operation, for more severe operating conditions and higher failure rate, the units of air blower non-plan outage or run reducing loads. The safety and reliability of operation of air blower directly affects the safe and economic operation steel mills and power plant. So, quickly determining the causes of failure and taking effective measures to solve it is a guarantee for continuous safe operation. The condition monitoring and fault diagnosis for air blower can improve the safety,reliability and effectiveness of the operation of air blower, improve the utilization of air blower, reduce maintenance time and human and material resources, improve maintenance efficiency, minimize costs and increase efficiency.This paper researches on monitoring and fault diagnosis technology of UHP EAF dust-removing air blower, which can realize monitoring and fault type diagnosis. On basis of many Domestic and abroad literatures, research status in quo and implementation process of monitoring and fault diagnosis of dust-removing air blower are expatiated; the treatment methods of monitoring signals and common signals are determined; with the structural characteristics of dust-removing air blower, status signal test programs of air blower are determined; According to the corresponding standard, monitoring threshold value of running state parameters of air blower are determined and condition monitoring systems of dust-removing air blower are established. Theoretical basis for fault diagnosis technology research are afforded through researching on the common faults of dust-removing air blower during running, analyzing failure mechanism and gaining the fault characteristic frequency. Typical fault samples are obtained by drawing eigenvector reflecting fault states of air blower. Fault diagnosis neural network are established with the characteristics of fault eigenvector and target vector of air blower; Neural network are trained with Typical fault samples of air blower and the fault recognization validity of fault diagnosis neural network are tested.