Advanced signal processing of tuyere pressure data to detect reduced hot wind throughput and raceway blockages
Sprache des Titels:
Proceedings of the 7TH European Coke and Ironmaking Congress - ECIC 2016
The present study deals with the analysis of blast furnace hot wind pressure data representing the wind throughput. In the frequent event of temporary tuyere blockages the flow rates for the affected tuyere section is reduced and is thus recognizable in the pressure signals.
In case when the blockages last for a longer time, additional fuel supplies like pulverized coal injection (PCI) must be switched off, to reduce the risk of damage on the tuyere system and blast furnace lining. This is usually implemented in the process control system by using a certain threshold level of the wind throughput. However, as the individual pressure transmitters tend to show a drift in their absolute values due to ageing and other effects, this approach is not able to react properly to all kinds of raceway blockages.
This study aims to find a more general approach of pressure signal analysis to provide a more precise detection of abnormal raceway behavior. The automatic signal processing of several hundred hours of blast furnace pressure data, allows to build a statistic and classification of various types of raceway blockage events. To check the significance of detected blockages the signal processing results are compared with tuyere video data.