The sintering of high-proportion iron ore concentrate is often plagued by such as poor permeability, inadequate granulation and insufficient liquid phase formation. The result showed that increasing the proportion of active quicklime could significantly enhance the sintering process. The problems mentioned above were systematically investigated and the mechanism by which a high-proportion of active quicklime improved sintering performance was explained in both theory and practice. Due to its high specific surface area and strong reactivity, active quicklime could quickly hydrate and release heat, thereby effectively suppressing the over-wet layer effect, increasing the mixture temperature, and optimizing the granulation structure. Simultaneously, it could promote the formation of key liquid phase minerals such as calcium ferrite and composite ferrites at lower temperatures, consequently improving the mechanical strength of the sinter. Based on sintering production practices at Ansteel's main plant, a process optimization path for applying a high-proportion of active quicklime was further proposed. The research results can provide a theoretical basis and technical guidance for stable and efficient sintering under high-proportion iron ore concentrate conditions, which is of great significance for advacing energy conservation, consumption reduction, and the upgrade of green metallurgical processes in the steel industry.

The annual production volume of converter slag in China is increasing, but 45% of it is still adopted by the traditional stockpiling method, which leads to soil pollution incidents and significant environmental and economic pressures. To adress these issues, internal recycling systems in domestic steel mills, such as the intelligent slag retention closed-loop control system, the gradient oxygen-supply dephosphorization process, hot metal pretreatment, and ladle slag supplementary materials were introduced. These technologies achieved a stable recovery rate of metal iron from converter slag at 80%~83%. However, the system still faces challenges such as compositional fluctuations (ω(CaO) ranging from 32% to 49%), high processing costs (85~120 yuan/t slag), and associated quality and safety risks. Future efforts should focus on advancing precision-oriented and cost-effective approaches to support the green transformation of the steel industry.

With the rapid development of machine learning, deep learning, and large model technologies, the intelligent empowerment of green steel manufacturing has become a research hot spot. The deep integration of digital and intelligent technologies with steelmaking process industrial scenarios was systematically explored, the current applications in key processes such as converters, electric arc furnaces, refining, and continuous casting were analyzed, the cutting-edge technologies such as multimodal perception and early warning, resource collaborative optimization, digital twins, and full-process causal reasoning were proposed to achieve green and low-carbon production as well as intelligent, efficient collaboration in steelmaking process. Finally, the challenges faced by the integrated and in-depth application of artificial intelligence in steelmaking-continuous casting section, potential strategies for large models to empower green manufacturing, and future prospects were discussed, which provided transformation ideas and theoretical support for intelligent empowerment of green steel manufacturing.

Precise detection was conducted on core parameters such as the surface micro-morphology, roughness, work-hardening degree, and magnetism of the strip steel after wet shot blasting descaling through systematic experiments. The results showed that there was no residual oxide on the strip steel surface, and the steel grit was not embeded into the strip steel substrate. Moreover, the surface state of the strip steel after descaling was quite consistent with that before descaling, which fully confirmed that this technology could endow the strip steel with excellent surface quality. In addition, based on the multi-combinations of different shot diameters and velocities, in-depth analysis of the descaling process was carried out using finite element simulation, revealing the internal mechanism of wet shot blasting descaling. The optimal descaling effect was achieved with a steel shot particle size (D) of 0.6 mm and a shot blasting speed (v) of 38.10 m/s. These findings provided important theoretical basis and technical guidance for the optimized application of acid-free descaling technology in actual production.

Slag sticking in ladles during smelting AGMJ45 steel is a key technical problem, which restricts the production of the steel. Experimental and theoretical methods were combined to determine the melting point and compositions of refining slag. Thermodynamic simulation software was used to analyze the liquid phase range of the refining slag and the relationship between its melting point and fluidity. The results showed that when the MgO content of the refining slag was from 4% to 7% and the basicity was from 4.0 to 6.0, the activity of Al₂O₃ was minimized while the activity of CaO was maximized, which enhanced the fluidity of the refining slag, formed the compounds with low melting point, and effectively improved slag sticking in ladles. Furthermore, modifications were made to the refining slag, and an optimized scheme for industrial production was proposed. After modification, the average melting point of the refining slag decreased by 123 ℃, and the amorphous rate exceeded 90%. Upon application, the average ladle weight at mid-campaign decreased by approximately 9.2%, and the average slag thickness decreased by approximately 23.4%. These improvements significantly alleviated slag sticking phenomenon, enhanced the continuity of ladle usage, and thereby increased production efficiency.

An analysis was conducted on the mechanisms and influencing factors of short scratch defects on the surface of cold rolling strip steel rolls during the grinding process. It was found that the short scratches on the roll surfaces, characterized as fine, intermittent linear marks, were closely related to grinding wheel vibration, improper characteristic matching and inadequate settings of grinding fluid parameters. Effective reduction of short scratches during the grinding process can be achieved through adjustments to grinding wheel, grinding fluid and grinding process parameters. These improvements not only enhanced the surface quality of the rolls but also increased grinding efficiency and the surface quality of downstream cold-rolled products. The short surfcace scratches were significantly reduced after implementing optimizations in the grinding process of rolls at a certain cold rolling mill demonstrated over a year and a half of practice, thereby substantially improving the quality of the steel plate surface.

To enhance the accuracy of wear prediction for coating layers on descaling rolls in critical metallurgical components and improve high-pressure water descaling efficiency, the application of machine learning in predicting the wear of the surface coating of the descaling rollser was deeply studied. The principles and structures of the prevalent algorithms of the multiple linear regression, the X G Boost and the random forest wear prediction model were introduced respectively. A data set was established using production data to predict wear coating on the surface of the descaling rolls. The predicted results were compared longitudinally with the measured ones, and the prediction results of the three algorithms were compared horizontally. The random forest wear prediction model achieved the highest prediction accuracy rates of 90.4% and 95.9% at error tolerances of ±0.001 0 mm and ±0.001 5 mm, respectively. Among the four performance indicators, the model achieved the highest coefficient of determination, while exhibiting the lowest values for root mean square error, mean square error, and mean absolute error in the other three indicators. The random forest wear prediction model proved optimal for actual production, providing high-value guidance for the setting of the deformation and reengineering empowerment mechanism of descaling rolls, which were key metallurgical components.

Aiming at the problems of 2600 m³ blast furnace in Benxi Steel, the main causes such as long-term furnace wall bonding, low water temperature difference, inactive bosh area and poor economic and technical indicators were analyzed. The causes were identified as the long-term excessive harmful elements in raw fuels, low bin level operation, and untimely adjustment of blast furnace operation. Insufficient blast volume, inadequate blast kinetic energy, hearth accumulation, severe furnace wall bonding and long-term unstable furnace condition were collectively induced by these factors. By adopting measures such as adjusting the charging system, optimizing the wall cooling system, improving the raw fuel quality, conducting manganese ore flushing, plugging tuyere, and adding coke for hot flushing of the hearth, the furnace condition gradually recovered to the normal production state.

Aiming at the problem that the nitrogen oxide emission concentration of the hot rolling reheating furnace in Chaoyang Iron & Steel Co., Ltd. of Ansteel Group Cor., Ltd. exceeded the standard and failed to meet the standard stably, the low-nitrogen combustion technology was formulated and applied combined with the actual production of the ASP hot rolling reheating furnace. On the basis of maintaining the heating capacity of the original burners unchanged, the top flat-flame burners and bottom heating adjustable-flame burners were replaced with low-nitrogen oxide environmental protection burners, and supporting technologies such as heating capacity configuration, combustion gas flow control, atmosphere control, and oxygen content monitoring were implemented. After application, the nitrogen oxide emission concentration was decreased from 182 mg/m³ to 93 mg/m³, and the fuel consumption was saved by 0.01 GJ/t. It not only reduced pollutant emissions, but also lowered energy consumption, achieving the goals of energy conservation, environmental protection and green carbon reduction.

According to the problems such as low production efficiency and high safety risks existing in the traditional control methods for boiler startup and shutdown in chemical production, APS (Automatic Procedure Starup/Shutdown) system control technology was studied and applied at oil burners of boiler in the whole process. The application results showed that the average temperature rise rate of the boiler was decreased from 160 ℃/h to 50 ℃/h, the startup time was decreased from 300 min to 215 min, the pressure fluctuation was dropped from 500 kPa to 200 kPa, and the input time of the oil gun was shortened from 240 min to 120 min. The fuel consumption for a single startup of the boiler was reduced from 18.81 tons to 13.43 tons and a decrease of approximately 28.6% was achieved, which improved the production efficiency and operational safety of the boiler and saved energy.