Drum Index Research Articles

Optuna-DFNN: An Optuna framework driven deep fuzzy neural network for predicting sintering performance in big data

Sintering performance index can reflect the quality of sintered ore, and the level of sintered ore performance directly affects the stability of blast furnace production. Accurate prediction of sinter performance metrics is essential to optimise sinter quality, improve production efficiency and reduce energy consumption. Based on the sintering big data, this paper proposes the Optuna-DFNN model by combining the fuzzy neural network algorithm, the deep neural network algorithm and the Optuna framework to address the problems of difficulty in tuning the parameter, lack of self-learning ability, and poor generalisation of the model of the traditional method when faced with the input of multiple parameters. The paper predicts the four prediction indexes of yield rate, return fines ratio, drum index and RDI+3.15 respectively. Among them, the relative error of the prediction of the yield rate is within 1.3 %, the relative error of the prediction of the return fines ratio is within 2.8 %, the relative error of the prediction of the drum index is within 0.32 %, and the relative error of the prediction of the RDI+3.15 is within 1.4 %. The results indicate that the Optuna-DFNN model has better performance in predicting sintering performance metrics. Based on the Optuna-DFNN sintered ore performance prediction model, the quality of sintered ore can be accurately predicted, the sintering process parameters can be adjusted in a timely manner, the sintering process can be optimized, and the utilization coefficient of sintering can be improved. Provide a reference for the optimisation of production processes, with a positive impact on cost reduction, environmental protection and increased sustainability of production.

Effect of ludwigite on sintering and metallurgical properties

Ludwigite is a multi-element coexisting iron ore. Through the sinter pot experiment, the effect of ludwigite on the metallurgical properties of the sinter is studied, and its action mechanism is revealed combined with X-ray diffraction (XRD) analysis. The appropriate amount of ludwigite is obtained on the evaluation mechanism of softening and melting properties of the sinter, which provides theoretical guidance for practical production. The results show that with the increase of the amount of ludwigite, the yield, vertical sintering speed, and drum index in the metallurgical properties of the sinter show a gradual downward trend, and the low-temperature reduction pulverization index RDI+3.15 shows a gradual upward trend. The addition of ludwigite is conducive to the formation of calcium ferrite and promotes the activity of MgO. The addition of excessive ludwigite will lead to the formation of Mg2B2O5 and Ca3B2O6, Slag phase is formed. With the increase of the proportion of ludwigite, the melting range of the sinter is slightly widened, and the permeability is improved first and then reduced. Under this ore blending condition, ludwigite resources can be added and used in sintering batching, and the proportion shall not exceed 6%.

Method to predict alloy yield based on multiple raw material conditions and a PSO-LSTM network

The production of ferroalloys accounts for a large proportion of the total energy consumption of the steelmaking industry. Accurately predicting alloy element yields is the key to reducing alloy waste, but there are significant differences in alloy yield under different conditions using ferroalloy raw materials during steelmaking. Therefore, this paper proposes a multi-model alloy element yield prediction method based on a particle swarm optimization (PSO) hyperparameter-optimized long short-term memory (LSTM) network and raw material condition classification. The accuracy of the PSO-LSTM prediction model was verified through simulations and was significantly higher than that of other network models when using the same raw material conditions. The average absolute error of predictions using raw materials with a low drum index was 0.4485, and it was 0.6162 under a high drum index, which was significantly lower than that (0.7077) under the condition without classification. This demonstrates the rationality of classifying working conditions according to the raw material conditions of ferroalloys. In addition, this paper combines the prediction model with a linear programming algorithm to develop a ferroalloy operating system and uses it in a steel plant to guide workers to complete an alloying operation. After four months of industrial testing, the internal control rate of finished steel composition increased from 91–94 % to 95–98 %. According to statistical analysis, the optimized HRB400E threaded steel consumed 1.23 kg less silicon-manganese per ton of steel, reduced the cost of steel alloy per ton by 8.6 yuan, and significantly reduced the waste of ferroalloys during steelmaking.

Prediction interval estimation of sinter drum index based on light gradient boosting machine and kernel density estimation

ABSTRACT Owing to the uncertainty operation in the sintering process, it is easy to produce uncertain prediction errors in the single drum index prediction model, which makes the prediction results lack certain reliability. Accurate and reliable prediction of the drum index can help improve the drum index. In this paper, a prediction interval estimation method of drum index based on a light gradient boosting machine (LightGBM) and kernel density estimation (KDE) is proposed. LightGBM can obtain accurate points prediction of drum index, and then use the KDE method to obtain the estimated prediction interval of drum index. The comparison results of different methods show that LightGBM has high prediction performance, and KDE can well quantify the prediction error of drum index, which verifies the effectiveness of the prediction interval estimation method combined with LightGBM and KDE, and provides more reliable decision-making information for the optimisation of sintering process parameters.

Research on Sinter Quality Prediction System Based on Granger Causality Analysis and Stacking Integration Algorithm

Sinter ore quality directly affects the stability of blast furnace production. In terms of both physical and chemical properties, the main indicators of sinter quality are the TFe content, alkalinity, and drum index. By analyzing the massive historical data on the sinter production of a steel company, this study proposes a sinter quality prediction system based on Granger causality analysis and a stacking integration algorithm. First, based on real historical data of sintering production in steel enterprises (including coal gas pressure, ignition temperature, combustion air pressure, etc.), data preprocessing of raw data was carried out using a combination of feature engineering and the sintering process. Second, Pearson correlation analysis, Spearman correlation analysis, and Granger causality analysis were used to screen out the characteristic parameters with a strong influence on the target variable of sinter quality (drum Index, TFe, alkalinity). Third, a prediction model for sinter quality parameters was developed using a stacking integration algorithm pair for training. Finally, a program development tool was used to realize the establishment and online operation of a sinter ore quality prediction system. The test results showed that the predicted goodness of fit of the model for the TFe content, alkalinity (R), and drum index were 0.942, 0.958, and 0.987, respectively, and the model calculation time met the actual production requirements. By establishing a suitable model and running the program online, the real-time prediction of the main indicators of sinter quality was realized to guide production promptly.

A Model Study on Raw Material Chemical Composition to Predict Sinter Quality Based on GA-RNN.

The quality control process for sintered ore is cumbersome and time- and money-consuming. When the assay results come out and the ratios are found to be faulty, the ratios cannot be changed in time, which will produce sintered ore of substandard quality, resulting in a waste of resources and environmental pollution. For the problem of lagging sinter detection results, Long Short-Term Memory and Genetic Algorithm-Recurrent Neural Networks prediction algorithms were used for comparative analysis, and the article used GA-RNN quality prediction model for prediction. Through correlation analysis, the chemical composition of the sintered raw material was determined as the input parameter and the physical and metallurgical properties of the sintered ore were determined as the output parameters, thus successfully establishing a GA-RNN-based sinter quality prediction model. Based on 150 sets of original data, 105 sets of data were selected as the training sample set and 45 sets of data were selected as the test sample set. The results obtained were compared to the real value with an average prediction error of 1.24% for the drum index, 0.92% for the low-temperature reduction chalking index (RDI), 0.95% for the reduction index (RI), 0.40% for the load softening temperature T10%, and 0.43% for the load softening temperature T40%, with all within the running time thresholds. The study of this model enables the prediction of the quality of sintered ore prior to sintering, thus improving the yield of sintered ore, increasing corporate efficiency, saving energy, and reducing environmental pollution.

Effect of Basicity on the Microstructure of Sinter and Its Application Based on Deep Learning.

The influence of the evolution rule of basicity (0.6∼2.4) on the mineral composition and microstructure of sinter is studied by using a polarizing microscope, and the comprehensive application analysis of the drum index, vertical sintering speed, and yield of sinter shows that, over the course of an increase in basicity (0.6∼1.0), the mineral structure changed from the original porphyritic-granular structure to a porphyritic structure. At the same time, there was no calcium ferrite phase in the bonding phase at a basicity of less than 1.0; therefore, the downward trend of the three indicators is obvious. When the basicity was further increased to approximately 1.6, the main structure of the mineral phase changed from a corrosion structure to an interweaving corrosion structure. Because of the existence of a porphyritic-granular structure, the structure of the mineral phase was extremely inhomogeneous and most complex near the basicity of 1.6; although a small amount of calcium ferrite displayed an acicular structure, the drum index appeared to show a very low value. With an increase in basicity to 2.0, the mineral phase structure was dominated by an interweaving corrosion structure with a uniform framework, and the content of calcium ferrite reached the highest value. Moreover, a clear acicular structure developed, and the drum index also increased to the highest value. At a basicity of more than 2.0, a mineral structure began to appear and the corrosion, porphyritic-granular structure, and the drum index also showed a slightly declining trend. Therefore, in the actual production process, basicity should be avoided as far as possible at around 1.0 and 1.6 and it should be controlled at around 2.0. At the same time, based on the mineral facies data set of this paper, the convolutional neural network is used to carry out a simple prediction model experiment on the basicity corresponding to the mineral facies photos, and the effect is quite good, which provides a new idea and method for the follow-up study of mineral facies.

Sintering Properties of Magnetic Separation Nonferrous Metallurgical Tailings

In this paper, the tailings produced by non-ferrous metallurgy were magnetically separated to obtain magnetic separation tailings. The sintering performance of magnetic separation tailings was studied by sintering cup experiment platform. In the research process, the differences in sintering process of different magnetic separation tailings were compared, including vertical sintering speed, sintering coefficient, drum strength, finished product rate, end point temperature and other sintering indexes. At the same time, the obtained sinter was subjected to chemical analysis to obtain its specific components. The influence of magnetic separation tailings with different contents on sintering was obtained by ore blending. The experimental results showed that the copper tailings are not sintered and the yield is only 0%. Tin tailings are not sintered and the yield is only 22.30%. The aluminum tailings can be sintered into lumps with a yield of only 57.10%. As the proportion of magnetic separation tailings increases, the vertical sintering speed decreases, the sintering utilization coefficient decreases, the sinter drum index decreases, the yield decreases, the sinter grade decreases, and the SiO2, S and P contents increase. In actual production, the proportion of magnetic separation tailings should be controlled below 10%.

The Study and Analysis of Sinter Ores` Fragmentation Situation after Times Continued Drum Test

The article concludes the drum index and wear index of sinter ores by times continued drum test. Based on the test results, we analysis the fragmentation situation of sinter ores and the principle which effects the grade distribution under drum test. The results show that after drum test, the quality of sinter ores between 0.00-5.00mm and 10.00-16.00mm rises apparently, while other part all get lower. What`s more, when the times of drum test are same, the powdery degree will get bigger if the time is longer. When the time of drum test is same, the powdery degree will get bigger if the times are more. When the times and times are all the same, the powdery degree will be closely, which shows the results fits the Experimental errors.

Production of high-strength and low-gasification reactivity coke from low-grade carbonaceous materials by vapor deposition of tar

Carbon/carbon composites were prepared from pyrolyzed-chars of lignite or low-strength coke (drum index of DI1506 = 52) and a coke-oven-gas tar through a gaseous-tar vapor deposition method to investigate the possibility of producing high-strength coke from low-grade coal or coke. The optimum vapor deposition temperature was examined under tar pyrolysis temperature of 700 °C and holding time of 30 min. The highest strength of prepared composite was observed at a vapor deposition temperature of 300 °C. The pore volume of the composite prepared at the vapor deposition temperature of 300 °C was the smallest among other samples prepared at that of 400–700 °C. The strength was improved by filling the carbonaceous materials derived from tar into the pores. The tar utilization coefficient was extremely low at 1.0 wt% or less. When same treatment was carried out for coke with different DI and pyrolyzed-cahr of lignite under the optimum conditions, it became clear that the composite prepared from any coke or pyrolyzed-char of lignite is comparable in strength to high-strength coke (DI1506 = 87). It was thus found that the present method can prepare high-strength coke from low-grade carbonaceous materials. The gasification rate of the composite prepared from low-strength coke was 0.5 times smaller than that of the original sample.

Effect of Coke Powder Ratio on Particle Emission and Sinter Quality in Sintering Process

Starting from the perspective of source reduction, the emission characteristics of particulate matters in sintering flue gas and sinter quality were studied by a sintering cup based on the fixed carbon 3.6% conditions in sintering raw materials. The particles were sampled by an Andersen 8-stage impactor at the coke powder ratios at 100%, 75%, 50% and 0%. The results showed that the particles in the sintering flue gas were mainly fine particles below 2.63 μm. When the coke powder ratio was 100%, the removal efficiency reached the maximum value of 34.37%. The effects of coke powder ratio on the sinter index were reflected in: when the coke powder ratio reduces, (1) the sintering speed decreases; (2) the solid fuel consumption increases; (3) the drum index improves. Compared with comprehensive indicators, the 100% coke powder was the most conducive condition to the production of enterprise sintering process.

Synthetically predicting the quality index of sinter using machine learning model

ABSTRACT A comprehensive prediction model of sinter quality based on machine learning algorithm was proposed. First of all, the mass historical data of actual sintering production was collected, cleaned and integrated. On this basis, the drum index and screening index of sinter were analysed by cluster analysis, and the quality of sinter was evaluated synthetically by clustering results and iron grade of sinter. Then, the important characteristic variables related to sinter quality index were screened by Recursive feature elimination, stability selection and random forest selection. The comprehensive classification model of sinter quality and the regression model of sinter’s total iron content were established by using various machine learning algorithms. The results show that the prediction accuracy of the classification model and regression model established by the extra tree is the best, and the application effect of the model is verified by using the testing set. The F1-score of the comprehensive quality index classification model is 0.92 and the R 2 of the total iron content regression model is 0.882. This model has good learning and generalization ability, and the accurate prediction of sinter quality index is realized.

Evolution profile of gases during coal carbonization and relationship between their amounts and the fluidity or coke strength

The evolution of inorganic-gases (H2, CO, CO2 or H2O) and hydrocarbon gases CH4, C2(C2H4, C2H6), or C3(C3H6 + C3H8) during the carbonization of caking coals is examined to clear the influence of the evolution amount of these gases on the Gieseler maximum fluidity (MF) and coke strength (DI15015). The samples are carbonized at 3 °C/min up to 1000 °C in high-purity He. The MF and the evolution profile of gases depend on the coal type. A negative relation exists between evolution amounts of gases, except for H2, until initial softening temperature (IST) and MF values. Additionally, the evolution amounts of inorganic gases, except for H2, until the IST increase with the increase of the carboxyl (COOH) group contents in raw coal. Moreover, the MF values trends to decrease with the increase in the COOH in raw samples. Thereby, it is found that the inherently-present coal-O adverse affect coal fluidity. Drum index (DI15015) of coke increases with the increase of C%-dry ash free (daf), fixed carbon (FC) %-dry, and reflectance (R0) vol% in raw coal, while the values decrease with the increase O%-daf and volatile matter (VM) %-dry. The DI15015 values exhibit an increasing trend with regard to the amount of the evolved H2 or CH4 during the carbonization. However, the values decrease as the amount of O-containing gases increased. A similar trend is observed in semi-coke and coke, where the values decrease as the remaining O% increased. Although the proportion of turbostatic-C (T-carbon) in coke depends on the coal type, DI15015 increases as the proportion of T-carbon increased. Therefore, it is found that the coal-O also has an adverse influence on coke strength.

Modified biomass fuel instead of coke for iron ore sintering

ABSTRACTOwing to majority of sintering fuel gas pollutants comes from the coke, modified charcoal replaced coke for iron ore sintering to achieve the emissions reduction of the pollutants in this study. By modifying biomass fuel with CaO and urea, the specific surface area, pore volume and pore diameter were reduced to 12.942 m2 g−1, 1.326e-02 cc g−1 and 18.865 nm. The experimental results of sintering cup showed that the vertical sintering speed increased, the burning through point advanced, the sinter drum index decreased gradually, and the sintering yield increased initially and then decreased with the increasing of modified charcoal content. The optimum proportion of modified charcoal instead of coke powder was determined to be 40%. Meanwhile, application of the 40% modified charcoal to sintering resulted in a decrease of 45.85% for SO2 and a decrease of 54.26% for NOx emissions.

Influence of tarry material deposition on low-strength cokes or pyrolyzed chars of low rank coals on the strength

In this fundamental laboratory scale investigation, carbon/carbon (C/C) composites were prepared from a mixture of pyrolyzed-chars of low-rank coals (lignite and sub-bituminous coals) or low-strength cokes (drum index: DI1506 = 52 and 73) and coke-oven gas (COG) tar to investigate the possibility of producing high-strength coke for blast furnace use from low-grade coals or cokes. In addition, the optimum conditions for composite preparation and the influence of the tarry material on the strength of the prepared C/C composite was examined. The C/C composites were prepared as follows: the tar was first added over particles of coke or char at room temperature, subsequently, the mixture was pyrolyzed at a predetermined temperature (500–900 °C) in an inert atmosphere. The indirect tensile strength of the C/C composite prepared at a pyrolysis temperature of 500 °C increased with increasing weight ratio of tar to cokes (3.5–4.5 MPa) or chars (2.0–3.0 MPa) up to a ratio of 2.0; at this point, the indirect tensile strength reached approximately 6.0 MPa. The indirect tensile strength of the C/C composites prepared with the weight ratio of tar to cokes/chars of 2.0 also increased with increasing pyrolysis temperature, reaching approximately 7.0–7.5 MPa at 800–900 °C, which was comparable to the ca. 7.0 MPa strength of the high-grade cokes (DI1506 = 87) used in commercial blast furnaces. Based on pore size distribution measurements using the mercury intrusion porosimetry, N2 adsorption, and polarized optical microscope methods, it was likely that the increase in the indirect tensile strength of the prepared C/C composite occurred because of the vapor infiltration of tar-derived carbonaceous materials into the micro-, meso- and macro-pores in the low-grade cokes and pyrolyzed chars during the pyrolysis of the mixtures.

Quality prediction modeling for sintered ores based on mechanism models of sintering and extreme learning machine based error compensation

Aiming at the difficulty in quality prediction of sintered ores, a hybrid prediction model is established based on mechanism models of sintering and time-weighted error compensation on the basis of the extreme learning machine (ELM). At first, mechanism models of drum index, total iron, and alkalinity are constructed according to the chemical reaction mechanism and conservation of matter in the sintering process. As the process is simplified in the mechanism models, these models are not able to describe high nonlinearity. Therefore, errors are inevitable. For this reason, the time-weighted ELM based error compensation model is established. Simulation results verify that the hybrid model has a high accuracy and can meet the requirement for industrial applications.

Burning loss reduction of sintering grates based on air flow simulation

ABSTRACTEdge burning of the 550 m2 sintering machine is a serious problem in Shougang Jingtang Iron and Steel Corp. To solve this problem, a lot of studies have been carried out, such as air flow simulation of the windbox and analyses on the grid section temperature and exhaust gas temperature. When 600 mm guide plates were added mounted on both sides of windboxes No. 15–No. 27 of sintering machine No.2 of Jingtang, the highest temperature decreased from 219–305°C to 201–227°C, and the temperature difference between the lower edge of the windbox and the central part shrank from 70°C to about only 30°C. After the optimisation, the service life of the grates increased from 3 months to more than 6 months, the drum index and energy consumption index of the sinter were improved, and the process energy consumption was reduced by 2 kgce t–1.

Online Heat Pattern Control of a Shaft Furnace Based on a Real-time Visualization by Particle Filter

In the steel works, direct observation of the internal states of many processes, such as the blast furnace, is difficult. Automation of such processes based on process visualization is an urgent issue. Because the number of sensors is limited, the state estimation utilizing partial sensor information is necessary. We developed a technique which visualizes the entire temperature distribution of a shaft furnace by means of the particle filter, which combines the sensor information and a nonlinear model calculation. This state estimation was incorporated in the heat pattern control logic based on future prediction, in which the estimated heat pattern is set as the initial condition. The control logic was implemented in a ferro-coke pilot plant. As a result, the control accuracy of 10°C was achieved. Furthermore, the operational condition was adjusted based on the correlation between the estimated heat pattern and the product strength. In consequence, the product strength improved by 0.5 points (Drum Index 150/15 mm, DI15015).

Characterization of Metallurgical Coke Produced with Coal Mixtures and Waste Tires

The use of scrap tires in the coal blends is an alternative to reduction of coke production costs as well as mitigates the impact caused by tire accumulation in the environment. Tests were carried out on a pilot scale, with coal blends and waste tire in 4 levels and 3 particle sizes, in order to assess the impact on the produced coke quality, especially on the Strength after Reaction with CO2 - CSR and Drum Index (DI150-15). In addition, it evaluated the mechanical resistance to compression, punctual chemical composition by Energy Dispersive Spectroscopy (EDS) analysis beyond the carbonaceous matrix-tire interface by SEM microscopy. The results from the coking tests showed that addition of up to 3% of the average ground tire (20-30mm) with steel mesh, raised either as CSR DI150-15, making feasible added also from a technical point of view.

Effect of Volume Breakage Due to DI Measurement on Pore Structure in Coke

Coke strength is an important property and generally evaluated with Drum Index (DI). It is considered that coke breakage occurs by two types of breakages during DI measurement, surface breakage and volume breakage. In this study, the cause of volume breakage was investigated by image analysis of cross-sectional image of coke before and after the DI measurement. As a result, a new image analysis method was developed. In this image analysis method, pore shape and coke-matrix connectivity were evaluated and new parameters “degree of Irregularity (I)” and “Connectivity Index (CI)” were proposed. I and CI represent pore shape and coke-matrix connectivity, respectively. By using this image analysis method, it was revealed that pore shape and coke-matrix connectivity were improved on adding HPC (High Performance Caking additive). By comparing the cross-sections before and after the DI measurement, it was found that the volume breakage occurred at the portions which have higher porosity, coarse pores, cracks and thin pore wall. Furthermore, it was clarified that the thickness of the pore wall broken by the DI measurement was < 7.6 µm thickness.