Subsurface hook formation during initial solidification in the continuous casting mould degrades the quality of steel slabs owing to the associated entrapment of argon bubbles and non-metallic inclusions. To minimise hook depth and to improve slab quality, extensive plant experiments were performed and analysed to quantify the effect of casting parameters on hook characteristics using the no. 2-1 caster at POSCO Gwangyang Works, Korea. The results reveal that meniscus heat flux plays an important role in controlling hook characteristics. Hook depth correlates with oscillation mark depth, hook shell thickness, and hook length. Based on regression analysis, this paper proposes an equation to predict hook depth in ultra-low-carbon steels as a function of casting speed, superheat, oscillation frequency, surface level fluctuations, and mould flux properties. Use of this quantitative equation enables improved control of subsurface quality in the continuous casting of steel slabs.

Spoken language understanding (SLU) aims to map a user's speech into a semantic frame. Since most of the previous works use the semantic structures for SLU, we verify that the structure is valuable even for noisy input. We apply a structured prediction method to SLU problem with comparison to unstructured one. In addition, we present a combined method to embed long-distance dependency between entities in a cascaded manner. On air travel data, we show that our approach improves performance over baseline models.Index Terms-Spoken language understanding, named entity recognition from speech, structured prediction, long-distance dependency, two-step cascaded approach

No abstract