Desulphurization is essential in the steelmaking process for high-quality steel production, and sulphide capacity has proven to be an effective index to evaluate the desulphurization ability of molten slag or flux. Several analytical or empirical models have been proposed to calculate the sulphide capacity. However, these models usually show insufficient generalization ability when new variables/data are introduced, which limits their practical application. In this work, experimental data were collected from the literature and a regularized extreme learning machine (RELM) model was established to predict the sulphide capacity of the CaO-SiO 2 -MgO-Al 2 O 3 slag system. The results demonstrated that the proposed model is robust for the prediction of sulphide capacity under different conditions. The coefficient of determination (R 2 ), correlation coefficient (r), rootmean-square error (RMSE) of the optimal model reached 0.9763, 0.9881, 0.113, respectively, which outperform the results of the reported models.
The temperature control of molten steel in ladle furnace (LF) has a critical impact on steelmaking production. In this work, production data were collected from a steelmaking plant and a hybrid model based on expert control and deep neural network (DNN) was established to predict the molten steel temperature in LF. In order to obtain the optimal DNN model, the trial and error method was used to determine the hyperparameters. And the optimal architecture of DNN model corresponds to the hidden layers of 4, hidden layer neurons of 35, iterations of 3 000, and learning rate of 0.2. Compared with the multiple linear regression model and the shallow neural network model, the DNN model exhibits stronger generalisation performance and higher accuracy. The coefficient of determination (R 2 ), correlation coefficient (r), mean square error (MSE), and root-mean-square error (RMSE) of the optimal DNN model reached 0.897, 0.947, 2.924, 1.710, respectively. Meanwhile, in the error scope of temperature from − 5 to 5°C, the hit ratio of the hybrid model acquired 99.4%. The results demonstrate that the proposed model is effective to predict temperature of molten steel in LF.
Intelligent iron/steel manufacturing has been drawing increasing attention in recent years. Slagging is of importance for clean steel production in the ladle furnace (LF) refining process. In this study, a calculation model of slag-making materials was established based on the actual LF refining conditions, the metallurgical mechanism model and the production data model of the sulphur distribution ratio. In order to test the slag-making model, the data of 50 heats from historical production were examined by using the model; as a result, 49 heats successfully reached the target sulphur content when the error of the lime weight is within the range of ±40 kg. Meanwhile, plant trials were carried out to validate the slag-making model; the results demonstrated that 45 out of 48 tested heats successfully reached the required level of desulphurization by adding slag-making materials at one time.
In order to consider both the refining efficiency of the ladle furnace (LF) and the quality of molten steel, the water model experiment is carried out. In this study, the single factor analysis, central composite design principle, response surface methodology, visual analysis of response surface, and multiobjective optimization are used to obtain the optimal arrangement scheme of argon blowing of LF, design the experimental scheme, establish the prediction models of mixing time (MT) and slag eye area (SEA), analyze the comprehensive effects of different factors on MT and SEA, and obtain the optimal process parameters, respectively. The results show that when the identical porous plug radial position is 0.6R and the separation angle is 135°, the mixing behavior is the best. Moreover, the optimized parameter combination is obtained based on the response surface model to simultaneously meet the requirements of short MT and small SEA in the LF refining process. Meanwhile, compared with the predicted values, the errors of MT and SEA for different conditions from the experimental values are 1.3% and 2.1%, 1.3% and 4.2%, 2.5% and 3.4%, respectively, which is beneficial to realizing the modeling of argon bottom blowing in the LF refining process and reducing the interference of human factors.
Based on the heat balance principle, and by using thermodynamic software FactSage6.4 simulation software, the study analyzed the impact of the ratio of iron tailings on the thermodynamic parameters of slag system, and explored the blast furnace slag and iron tailings and the change
rule of heat transfer between the blast furnace slag and iron tailings and the influence law of the ratio of iron tailings on the temperature and concurrent heating quantity of the slag system; the study further adopted the least squares method, multiple regression analysis method to establish
the mathematical model of the temperature drop ΔT and concurrent heating quantity of the slag system; Based on the above research, the study constructed the coordinated control chart of the ratio of iron tailings, the temperature of slag fluidity, and concurrent heating quantity
so that the law of concurrent heating of the slag system by taking the iron tailings as conditioning agent was obtained.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.