Detection method of molten iron flow velocity at blast furnace taphole combining visual perception and jet mechanism

Online detection of molten iron flow velocity at the blast furnace taphole is of great significance to solve abnormal tapping problems in time. However, the existing detection methods have the problems of low accuracy or high computational complexity. To this end, this paper proposes a detection method of molten iron flow velocity by combining visual perception and jet dynamics mechanism. Firstly, a velocity detection system is designed to acquire visible images of the molten iron flow. Then, a method based on two-stage Hough transform (HT) is proposed to position the boundary curve of the flow, so as to extract its morphological features. Next, a jet dynamics model is established to quantitatively depict the mapping relationship between molten iron flow morphological features and the velocity, thus the velocity can be calculated directly in a single frame image. Finally, industrial experiment results show that the proposed method can detect the velocity accurately.