Design Exploration to Determine Process Parameters of Ladle Refining for an Industrial Application. Issue 10 (15th March 2016)

• Record Type: Journal Article
• Authors: Shukla, Rishabh; Anapagaddi, Ravikiran; Singh, Amarendra K.; Panchal, Jitesh H.; Mistree, Farrokh; Allen, Janet K.
• Is Part Of: Steel research international. Volume 87:Issue 10(2016)
• Page Start: 1333
• Page End: 1343
• Publication Date: 2016-03-15
• Subjects: ladle refining -- response surface models -- compromise decision support problem -- ternary plots; Steel -- Periodicals; Steel -- Metallurgy -- Periodicals; 669.142
• DOI: 10.1002/srin.201500392
• Languages: English
• ISSNs: 1611-3683

Abstract

Abstract : Due to increasing competition and the demand for light‐weight steel, manufacturers strive to produce new grades of steel to meet stringent sets of property specifications. The authors are informed by engineers at a major steel‐making plant that for them to determine the set‐points for the ladle, tundish, and caster to make a new grade of steel it takes between 20 and 25 physical plant trials. These plant trials are expensive and very difficult to schedule. In this paper, the authors focus on explicating the method for one unit operation in the steel‐making process, namely, ladle refining, which is a key operation for maintaining cleanliness and controlling composition; these have a significant bearing on the steel's mechanical properties. The challenge is to determine set points for ladle refining using computational models. In this paper, the authors present a method for visualizing and exploring the solution space using the compromise Decision Support Problem (cDSP). To illustrate the efficacy of the method, the authors determine the set points of the ladle operation in an industrial setting. The focus of the paper is to elucidate the utility of method and not on the analysis models used in the framework. Abstract : In industry, ladle refining of steel has a significant impact on the cleanliness and composition of advanced high‐strength steel. A model‐based design framework is developed to explore the process design space and predict set points to meet given requirements. The method is useful for designing a process with fewer plant trials and can be instantiated for other unit operations, which in turn, helps to reduce the cost and time associated with scaling production from laboratory to industrial scale.