Vibration style ladle slag detection method based on discrete wavelet decomposition

Abstract

The purity of molten steel is a crucial factor of steel products' quality. Detection and removal of slag carryover from the molten metal is an important task in steel making procedure, especially in the continuous casting process. Researchers have applied several different slag carryover detection methods to solve the slag carry-over since 1980s, such as electromagnetic coils method of AMEPA, Infrared thermal infrared imaging method, supersonic wave method and vibration method, which are all non-contact methods and some other contact type slag detection methods. In this paper, a non-contact vibration analysis slag detection method is presented. The main principle of this method is based on the vibration variety of ladle shroud between pure molten steel fluid and slag carryover fluid. The experiences of steel making operators have proven the differences of vibration wave in continuous casting process. The signal data gathered from produce field is analyzed by power spectral density (PSD) and discrete wavelet decomposition (DWD). The experiment results are given, the vibration wave feature of slag carryover is distinctive. Which verified the validity of vibration slag detection method.