BOF Process Control and Slopping Prediction Based on Multivariate Data Analysis

Self Cite

In the Basic Oxygen Steelmaking (BOS) process, a heterogeneous emulsion‐solid mix will form, consisting of an emulsion of liquid slag and metal droplets, in which 2nd phase particles of undissolved fluxes and solid in‐blow precipitates are suspended. When the carbon in the metal droplets reacts with iron oxide, small bubbles of CO gas are formed. If the upward movement of these bubbles is obstructed by the physical properties of the emulsion‐solid mix, foaming will occur. Certain process conditions may lead to an excessive foam growth, in the worst case forcing foam out of the vessel. This undesired process event is known as “slopping”. Extensive studies during recent decades have shown that emulsion characteristics strongly connected to foaming are: viscosity, surface tension, and density. The extent of foaming is also dependent on bubble size; foaming increasing with smaller bubble size. However, investigations into the influence of the mineralogy and morphology of the emulsion‐solid mix on foaming in basic oxygen steelmaking are scarce. In this work, samples from trials in a 6‐tonne pilot plant BOS vessel are examined by XRD and with SEM for the determination of emulsion‐solid mix mineralogy and morphology at different stages of the oxygen blow. The study confirms the importance of tight process control in order to minimize the emulsion‐solid mix apparent viscosity and, hence, the foam height, but this without over‐oxidizing the liquid slag phase, which would result in increased gas generation within the slag‐metal emulsion.

Section: Resultsmentioning

confidence: 99%

Characterization of Slag‐Metal Emulsion and Its Impact on Foaming Behavior and Slopping in the LD Process

Brämming

Engström

Samuelsson

et al. 2018

Self Cite

“…The slopping tendency as measured by the increase in the oxygen lance vibration is the main controlling factor because this parameter is directly associated with the undesired emission phenomena . The imminent slopping was revealed by the expert system mounted on the oxygen lance of the BOF, giving operators a real time warning of the onset of slop and its severity.…”

Section: Discussion Of the Experimental Testing Resultsmentioning

confidence: 99%

Study of a New Operating Practice for Refining High Silicon Hot Metal in a BOF Converter

Barella

Cecca

Mapelli

et al. 2015

Within the context of Basic Oxygen Steelmaking, the refinement of high silicon hot metal is a cumbersome operation due to the strong tendency for slopping as the silicon concentration in the hot metal increases. According to generally recognized steel working practices, this issue could be resolved by a double-deslagging (slag removal) operation. In the present study, a new operating practice has been defined that eliminates slopping and increases productivity. The new tested procedure is a single-slag operation that results in a reduction in the tap to -tap time and greater gas recovery, affecting neither the slopping control nor the quality of the steel obtained at the end of the oxygen blow. The feasibility of such a new practice has been verified through a critical analysis of the results from thirty 30 experimental tests carried out at the industrial converters of a steel plant

“…For instance, to monitor the height of foam slag in converter smelting process, SSAB Lulia Steel Works in Europe studied a measurement method on vibration of converter furnace by applying multivariate data analysis and discussed the practical application in predicting end phosphorus content in molten steel. [ 1 ] Similarly, to instruct the converter smelting process, Bae and others used standard machine‐learning model to predict converter molten steel end temperature, carbon content, and phosphorus content based on data feature engineering approach. [ 2 ] Han and others designed a prediction model for converter end parameters with anti‐interference ability based on method combining evolutionary membrane algorithm and extreme learning machine.…”

Section: Literature Reviewmentioning

confidence: 99%

An Integration Model for Converter Molten Steel End Temperature Prediction Based on Bayesian Formula

Feng

Yang

et al. 2021

Highly accurate prediction of converter molten steel end temperature is important foundation of realizing intelligent smelting in converter procedure. Single model prediction of converter end temperature has problems such as weak generalization ability and difficulty in increasing accuracy. To tackle these problems, this review proposes a modeling method based on Bayesian formula that dynamically integrate multiple models with different features. First, various data patterns and features are explored using different modeling methods and respective prediction model for converter molten steel end temperature is constructed. Then, the value range of the objective is discretized and the confidence levels of the prediction results from different models are computed based on Bayesian formula. Last, the prediction values of different models are weighted according to their confidence level and a comprehensive prediction result is obtained. Testing and comparison are carried out on the integration model and the three single models (support vector regression, random forest, and BP neural network) using actual production data. Results show that the integration model based on Bayesian formula can further increase the prediction accuracy effectively on top of single model prediction accuracy.