Stainless steel/carbon steel clad plate as an important structural material was widely applicated in petrochemical industry, water conservancy and hydropower and ocean engineering for its excellent strength, ductility and corrosion resistance. However, the material was faced with the problem of complex and difficult welding process. This paper aims at improving the welding application of stainless steel/carbon steel clad plate. A double-sided laser welding process was developed in 2 mm +6 mm thickness stainless steel/carbon steel single-sided clad plate, in which stainless steel and carbon steel were self-fusion welded by laser with high energy density and accurate welding depth. As a result, the welding process was simplified and excellent performance in welding joint was achieved. The results indicated that the designed welding process had a good weld-ability, and a welding joint with small weld pool, small heat affected zone, uniform weld and small deformation in plate. Also, the corrosion resistance performance of the stainless steel side welds was tested and analyzed in detail. The good corrosion resistance was obtained in stainless steel layer.

This paper examines the evolving trends in the digital transformation of the steel industry and provides a systematic overview of the research and engineering achievements of the State Key Laboratory of Digital Steel (DSL) in the domain of steel rolling. A series of high-fidelity digital twin models integrating data, physical mechanisms and empirical domain knowledge have been developed for critical production processes, including hot rolling, cold rolling and heat treatment. Intelligent optimization methods tailored for industrial production environments have been designed and implemented in large-scale steel enterprises both domestically and internationally. These innovations have yielded significant improvements in quality metrics, particularly in three-dimensional dimensional accuracy, microstructure, and mechanical properties. Finally, the paper outlines the future research and application prospects of "AI + steel rolling".

Combining the calculation of resistance to deformation of rolled piece with the based finite element calculation of elastic deformation of roll system, the model of shape control in rollingsilicon steel thin-gauge strip with Sendzimir 20-high mill was established.By calculating the deformation resistance of the rolled piece based on the preset shape and iteratively calculating the elastic deformation between the roll system under the rolling force of the rolling mill, the influence laws of the support roll gap shape adjustment (ASU) and the taper change of the No.1 intermediate cone roll on the shape of the rolled piece were determined.The adjustment of ASU in the middle of the support roll mainly affects the area of 0.6 m at the center of the width of the rolled piece, and has almost no effect on the area of 0-0.2 m at the edge. The adjustment of ASU at both ends of the support roll has an impact on the edges of the rolled piece, and also on the middle part of the rolled piece. Increasing the taper of the intermediate roll has a significant effect on improving the strip crown at a distance of 0-0.15 m from the edge of the rolled piece. By optimizing the calculation of the ASU of the roll system and the axial shifting amount of the intermediate roll, the average difference of transverse of the same silicon steel thin-gauge strip rolled by Sendzmir 20-high is reduced by more than 25%,and the shape control of the 0.25 mm thickness strip reaches below 8 IU, effectively improving the shape control capability of the Sendzimir 20-high mill for silicon steel thin-gauge strip.

The development history of heavy plate production in China was sorted out, and the development of heavy plate in China was divided into five stages, i. e, initial stage, accumulation stage, development stage, maturity stage and optimization stage, The characteristics of each stage were analyzed from the aspects of mill specification, equipment level, capacity scale and so on; The technical progress and development of key processes and equipment for heavy plate, such as hot delivery and hot charging, reheating furnace, rolling mill and leveler were described; The development, application and advancement of typical heavy plate products, such as special shipbuilding steel and offshore engineering steel were elucidated; The future development of heavy plate in China was prospected and suggestions were put forward.

In recent years, China's rolling technology has rapidly advanced from being a follower to a parallel and then a leader. In terms of processes and equipment, China has achieved international leading levels in a series of innovations and applications, including multimode headless continuous casting and rolling technology, 5 600 mm wide-heavy plate rolling technology, 2 680 mm wide stainless steel hot continuous rolling production lines, six-stands six-high tandem unit for high-grade silicon steel, high-precision heat treatment processes and equipment for strip and plate, key production technologies for ultra-thin stainless steel sheets, and online microstructure and property control technology for hot-rolled seamless pipes. In steel products, China's advanced steel enterprises have essentially reached international leading standards in the quality of automotive steel, energy steel, engineering machinery steel, shipbuilding and offshore engineering steel, low-cost 2 000 MPa-level ultra-high-strength and ductile steel, and certain specialty steels. In terms of the "strength-plasticity-toughness" control mechanism for ultra-high-strength steel, innovative approaches such as the "dislocation engineering + TRIP effect" synergy mechanism and the "martensite topology design + metastable phase regulation" synergy plasticity enhancement mechanism have been proposed. Meanwhile, China has also reached international advanced levels in digital and intelligent empowerment for high-quality and stable rolling production.Overall, China has made significant progress in rolling processes, equipment, products, digitalization, and intelligent empowerment, but there remains a gap compared to international advanced standards in core rolling processes and equipment, large rolling mill control systems, and high-end, high-quality product technological innovation. Some core equipment and control systems still require imports.Considering the development trends of China's steel industry, key technological innovations and development directions to focus on include: core rolling process and equipment technology, high-end and high-quality product innovation, continuous and efficient key technologies integrated with upstream/downstream processes and the industrial chain, as well as digital and intelligent empowerment.

Aiming at the challenge of integrated control of key indicators such as shape, properties, and surface quality in the production of high-quality sheet and strip, the achievements of China in the integrated control technology of shape-property-surface for high-quality sheet and strip were introduced. The strip shape control technologies of multi-modal information fusion detection and multi-stand and multi-process coordination were briefly described from the aspects of straightness detection of hot rolled strip, intelligent prediction system of roll and strip shape state, comprehensive control model of roll bending and axial shifting, and strip shape feedforward control of cold and hot rolling process. The control technologies of hot rolling process and mechanical properties based on large model were briefly described from the aspects of microstructure parameters and mechanical properties control model of hot rollingprocess and pre-control of microstructure properties of hot continuous strip rolling. The intelligent simulation and control technology of high-quality strip surface characteristics was briefly described from the aspects of unsupervised classification detection algorithm for surface defects and comprehensive control of surface defects and characteristics. On this basis, the field applications of these technical achievements were described, and the development of comprehensive control of high-quality sheent and strip was prospected.

Nickel (Ni) is an important alloying element widely used in steel, which not only significantly enhances the corrosion resistance and heat resistance of stainless steel, but also improves the strength, ductility, and toughness of alloy steel. Nickel-based alloys are also applied in the manufacturing of high-temperature aerospace structural components due to their high strength and good heat resistance. Although nickel has been fully utilized in steel and alloys, especially in improving the toughness/cryogenic toughness of steels, the physical metallurgical mechanism of nickel in enhancing the toughness of steel has not been fully understood. In this paper, the application and research progress of nickel in steel and alloys are first briefly reviewed, and then the mechanism study of nickel in improving the impact toughness and welding toughness of steel are summarized. It is emphasized that the improvement of impact toughness of steel by nickel is mainly achieved by grain refinement, dislocation movement promotion, increment of high-angle grain boundaries, and stabilizing austenite, while the welding toughness is enhanced by effectively increasing the content of acicular ferrite and reducing the harmful phase of martensite-austenite islands (M/A). As a new approach to strengthening and toughening steel, the research status of Ni-Cu alloying in low-alloy high-strength steel are briefly described. Finally, the development trend of Ni-containing steel is provided.

Front-end warping and bending of slab during hot roughing rolling is a common asymmetric shape defect. Traditional control methods typically adopt an integral control strategy for the slab head, which may result in excessive warping, leading to slab collisions with production line equipment and disruptions in rolling rhythm. To address this issue, a segmented control method for slab front-end warping and bending was developed and validated through finite element simulations. The proposed method consists of two stages,i.e. the sled coefficient action stage and the upper/lower roll speed reversal stage. Using machine vision technology, the amount of warping or bending between passes is detected, and the corresponding adjustment value is calculated through the segmented control strategy in combination with the front-end warping and bending monitoring and control system. The adjustment command is then automatically applied before the next rolling pass, thereby realizing automatic control of slab front-end warping and bending. Application results demonstrate that the segmented control method reduces the slab head warping height by 77.04% compared with integral control, significantly mitigating front-end warping and bending defects and enhancing the stability of product quality.

During the rolling process of medium and heavy plates, various types of plane shape defects may occur, which severely affect product quality and yield, thereby constraining performance and production line efficiency. As a critical technical aspect for ensuring the dimensional quality of final products, planar shape control has long been a key focus in the field of steel rolling. The development of planar shape control technologies for medium and heavy plates are systematically reviewed, covering the evolution from fundamental theories and experimental studies to engineering applications, with concise analysis, comparison, and commentary. Furthermore, recent advances in intelligent equipment and data-driven control technologies are summarized, and future research directions are discussed in the context of the steel industry's ongoing shift toward intelligent manufacturing. The aim is to provide theoretical support and practical insights for the continued optimization and upgrading of planar shape control technologies.

In the production process of hot-rolled strip, the control accuracy of finishing rolling temperature directly affects the microstructure and mechanical properties of the strip, which is a key factor in ensuring dimensional accuracy and good flatness of the strip. Due to the complexity, multivariability and strong coupling characteristics of rolling process, traditional mechanism models have obvious deficiencies in prediction accuracy, which is difficult to meet the requirements for high-precision and high-performance products control.To this end, this article combines a domestic 2 250 mm hot strip mill with the GBDT algorithm and mechanism model to develop a predictive model for the finishing rolling temperature of hot-rolled strips that integrates both mechanistic models and data.This model has both the physical interpretability of a mechanism model and the ability to fully develop GBDT algorithm's advantages in data mining, enabling continuous learning and adjustment of historical data and real-time feedback, thus maintaining the stability and reliability of the model's predictive performance;At the same time, it has self-training and closed loop control functions to achieve automatic closed loop control.After applying the model online, the deviation of long-term genetic prediction for finishing rolling temperature has decreased from 14.55 ℃ to 9.85 ℃. The results show that the model has high calculation accuracy and can meet the requirements for finishing rolling temperature control under different steel grades and working conditions, thereby improving the stability of strip rolling and the precision of head finishing rolling temperature control, enhancing product competitiveness.

Carbon capture, utilization, and storage (CCUS) technology, as the only pathway to near-zero emissions of fossil energy at present, is crucial for reducing CO₂ emissions and supporting China′s "dual carbon" strategic goals. The large-scale, long-distance CO₂ pipelines, which connect CO₂ capture and storage sites, play decisive roles in ensuring the efficient operation of CCUS projects. In this work, the construction status of important domestic and international CO₂ pipeline projects is summarized basing on the phase characteristics and transport status of CO₂. The influence mechanisms of water content, impurity gas, and transportation technology on the corrosion behavior of CO₂ pipelines are summarized. The research progress on corrosion prevention methods such as corrosion-resistant pipeline materials, coatings, and corrosion inhibitors is reviewed. The future development direction of CO₂ pipelines is prospected. Compared with the demand of CCUS project and the mature CO₂ transportation pipeline technology in the world, there are certain gaps in the construction and experience of long-distance and large-scale CO₂ pipelines in our country, and it is urgent to deepen the engineering practice research of supercritical CO₂ pipeline transportation technology. At present, most studies on CO₂ pipelines are concentrated in the laboratory environment, and there are discrepancies in some research findings, which cannot effectively guide the construction and practical engineering application of CO₂ pipeline in China. Thus, it is urgent to establish pilot-scale experiment platforms and carry out massive experimental studies to reveal the interaction mechanism of multiple factors under actual complex working conditions. In addition, it is suggested to use cost-effective bimetallic composite pipes and develop novel types of efficient corrosion resistant coatings and corrosion inhibitors suitable for bimetallic composite pipes, in order to form a complete set of new protection technology for CO₂ pipelines.

Aiming at the problem of excessive manual intervention of the existing cold rolling finishing unit, the control system and core database of the existing finishing unit were analyzed, and the technical scheme of automatic control and upgrading of finishing unit was formed. Secondly, the alternating control strategy of constant elongation and constant rolling force mode was determined, the automatic function of logic control and execution control module were improved, and the core database of the control system was expanded and refined. Finally, the process adjustment program of constant elongation and constant rolling force was optimized. The feed-forward preset parameters and the flatness feedback closed-loop control and adjustment mechanism of the finishing unit were established, and the automatic control function of the plate shape of the finishing unit was completed. This research is applied to the industrial production, and the automatic control function of the finishing unit based on the online data is realized. The automation rate of the unit increases to more than 90%, the manual operation and the scrap rate are reduced.

The rolling force prediction model is the core of the cold rolling set control system. For a long time, the traditional rolling force theoretical model has exhibited low prediction accuracy and a strong dependence on empirical parameters due to the complex influencing factors in the cold rolling process, such as multivariable interactions, strong coupling, nonlinearity, and time-dependence. These limitations have hindered its ability to meet the production requirements of high-precision cold-rolled ultra-thin-gauge strips. The setting of the rolling force primarily depends on the calculation accuracy of deformation resistance and the friction coefficient. By analyzing the classical Bland-Ford-Hill cold rolling force theoretical model, inverse calculation formulas for deformation resistance and the friction coefficient were established, thereby obtaining their true values. Subsequently, a least squares support vector machine (LSSVM) model optimized by the differential evolution (DE) algorithm (DE-LSSVM) was constructed. By inputting the true values of deformation resistance and the friction coefficient into the DE-LSSVM for training, corrections were made to these parameters, thus optimizing the rolling force theoretical prediction model. Experimental results demonstrate that, compared with the traditional rolling force theoretical model, the deviation of the rolling force prediction model based on optimization of deformation resistance and friction coefficient is controlled within 5%.

The roll gap calculation model of L₂ system is the foundation for controlling the thickness of hot rolled medium and heavy plates. Especially when employing the feedback pressure AGC control, the accuracy of roll gap calculation model of L₂ system directly affects the thickness of the head and the whole plate, consequently affecting the yield of medium and heavy plates. A 2 680 mm semi-continuous roughing rolling mill utilizes the pressure feedback AGC control system, but faces the problem of low thickness accuracy in rolling stainless steel medium and heavy plates. To address this problem, a multi-factor regression analysis was conducted using process data from the intermediate billet rolling process and actual thickness data measured by the thickness gauge. Sixteen indicators were selected for analysis to determine the correlation and path analysis of the thickness of the intermediate billet in cold state. The analysis results indicate a collinear effect of various factors on the thickness of the billet in cold state. The main influencing factors identified in the process of rolling medium and heavy plates are the roll gap, number of rolled pieces per unit of the roll, and slab discharge temperature. A multivariate fitting regression model was developed based on these key influencing factors to correct the roll gap of the R₇ stand during the rolling process. Extensive production practices have demonstrated that the application of the fitting regression model can significantly enhance the setting accuracy of the R₇ stand roll gap. This improvement has led to an increase in the overall thickness accuracy rate of medium and heavy plates from 84.47% to 98.92%.

Crown is one of the critical quality indicators for medium and heavy plate products, and its control level is a testament to an enterprise's production capacity and market competitiveness. With the vigorous development of artificial intelligence (AI) technology, intelligent control has become the core strategy driving the development of the iron and steel industry. However, there is still a lack of relevant research on how to apply AI technologies such as big data to improve product quality, especially in enhancing the plate crown control level during the medium and heavy plate rolling process. This study focuses on the plate crown control technology of medium and heavy plate mills. By combining mechanism models with data-driven models, an intelligent prediction model for heavy plate crown is established, achieving a hit rate of 83.04% for plate crown within the range of ±80 μm. Meanwhile, based on the theoretical framework of this model and big data analysis technology, the roll contour curve of the finishing mill work rolls is optimized. This optimization not only improves the crown control capability of medium and heavy plate mill products, but also extends the rolling cycle of work rolls, and increases both the mill productivity and product yield.

Aiming at the problems in the current production of SWRH82B wire rods, including poor overall performance along the entire length, abnormal structures (such as high-level network cementite and martensite), low and uneven sorbitization rate, this paper conducted a finite element simulation calculation of the temperature field throughout the entire rolling process of SWRH82B wire rods using DEFORM-3D software, based on a high-speed wire rod production line of a steel plant. The influence of plastic deformation of the workpiece on heat exchange between its surface and core during rolling was analyzed. Meanwhile, the optimization of water cooling processes was investigated.The results show that: In the roughing stage, the surface temperature of the workpiece shows a gradual decreasing trend, while the core temperature decreases slowly; in the intermediate rolling stage, the surface temperature rises, and the core temperature decreases more significantly. During the pre-finishing stage, the cross-sectional area of the workpiece gradually decreases, and the entire section heats up due to heat generation from plastic deformation, leading to increased temperatures in both the surface and core. In the finishing and reducing-sizing stages, with higher rolling speeds, the surface and core temperatures of the workpiece briefly increase, then rapidly decrease after water cooling, followed by a rise in surface temperature due to core reheating. When the convective heat transfer coefficients of No.1 to No.4 water cooling boxes and No.5 water cooling box are 400 W/(m²·℃) and 300 W/(m²·℃), respectively, the average surface temperature of the wire rod at the laying head is approximately 919 ℃, and the core temperature is about 939 ℃.After optimizing the water cooling process, the cross-sectional temperature of the workpiece becomes uniform, with the simulated temperatures basically agreeing with the measured ones (error within ±10 ℃). Meanwhile, the martensitic structure in the core of the wire rod is significantly reduced. Compared to before optimization, the tensile strength of the wire rod increases by approximately 40 MPa, and the reduction of area increases by about 8%.

In the production process of medium and heavy plates, the edge parts of plates often present irregular shapes. These irregular areas are generally unsuitable for further processing or application and need to be accurately cut off. To improve shearing efficiency and accuracy, an intelligent cut-to-length technology for segmented shearing of medium and heavy plates, which combines machine vision and PLC control, has been proposed.Firstly, the camera is accurately calibrated using Zhang Zhengyou's calibration method, and the spatial positional relationship between the camera and the shearing machine is determined. Then, the RGB color values of the image are extracted and further converted to the HSV color space, so that the image features are effectively extracted, and thus the endpoints of the plate head are accurately identified. Moreover, the distance between the endpoints in the world coordinate system is calculated by means of pixel coordinate conversion.When it is detected that the width of the plate head is stable and the number of endpoints is two, the system automatically determines that it is the regular position of the plate head. Subsequently, the system calculates the distance from the edge point of the length-measuring laser line to the regular position, so as to determine the target shearing position. The target position is sent to the PLC control system, and the roller table is controlled by the PLC to complete the shearing action.This intelligent fixed-length technology not only improves the accuracy of plate end-point recognition,but also significantly enhances the shearing accuracy and product quality, with the fixed-length accuracy controlled within ±5 mm.

Steel materials elongation, as a core indicator for characterizing material plasticity, directly affects the formability and service safety of steel products. To address the problems of insufficient generalization across steel grades, lack of mechanism support, and dependence on destructive experiments in existing elongation prediction models, a mechanism-data fusion based on Bayesian optimization-LightGBM prediction model is proposed. Based on industrial production data, a multidimensional input system of "conventional process/composition characteristics+mechanism characteristics" is constructed, where the mechanism characteristics include carbon equivalent (C_eq), welding crack sensitivity coefficient (P_cm), and downstream pass reduction rate; a Bayesian optimization algorithm is adopted to efficiently search for the key hyperparameters of LightGBM; finally, the performance of the model is verified through ablation experiments and comparative experiments with multiple models. The experimental results show that the coefficient of determination (R²) of the proposed model reaches 0.898 4,the hit rate with an error within ±3% is 96.6%, which provides reliable technical support for quality control in steel materials production processes.

Aiming at the problems that HRB400E reinforced bars have a relatively high manganese (Mn) content and the finished products are prone to surface red rust when adopting the traditional process of high-temperature finishing rolling plus water quenching cooling after rolling, this paper optimized the temperature regime of the rolling process specifically, controlling the rolling start temperature at 960-1 000 ℃, the temperature entering the finishing rolling mills at 830-850 ℃, the temperature entering the reducing mill at 830-850 ℃, and eliminating water quenching cooling before the bars are sent to the cooling bed, and the corresponding production experiments are carried out.The results show that the produced 14 mm HRB400E reinforced bars achieve a grain size of Grade 11, a yield strength of 452 MPa, a tensile strength of 655 MPa, and a total elongation at maximum force of 14.5%. Moreover, the iron oxide scale on the bar surface is dense and intact without red rust, and the mass fraction of Mn can be reduced by 0.1%. This study provides a basis for the green and low-carbon production of HRB400E reinforced bars as well as the improvement of their mechanical properties and surface quality.

During strip cold rolling, tension can alter the deformation resistance of the strip, thereby inducing fluctuations in thickness of strip; whereas variations in strip reduction, in turn, trigger changes in strip tension. Under the mutual interaction between tension and thickness, tension-thickness coupled fluctuations arise. Such fluctuations adversely affect both the thickness accuracy of the product and the stable operation of the mill. Focusing on the frequent occurrence of tension-thickness coupled fluctuations during thin-gauge strip rolling on a Sendzimir 20-high reversing cold rolling mill, this study investigates the structural principles of thickness control, tension control, and the inertial compensation function of Automatic Gauge Control (AGC). As demonstrated in this study, sudden fluctuations in the inlet strip linear speed are identified as the root cause of tension-induced elastic oscillations.The interaction between these elastic oscillations and AGC roll gap adjustments gives rise to tension-thickness coupled fluctuations; consequently, the inertial torque compensation adjustment of AGC must remain fully effective throughout the entire rolling process.To address this requirement, the calculation procedure for AGC inertial compensation torque is deduced, and the factors contributing to the failure of AGC inertial torque compensation are analyzed. By optimizing the torque dead-zone parameters of the uncoiling drive and enhancing the tension setting during the tail rolling stage of thin-gauge strip, the effectiveness of AGC inertial torque compensation is maintained throughout the rolling process, and tension-thickness coupled fluctuations are eliminated. Furthermore, a variable-speed integral control algorithm is employed to suppress tension-induced elastic oscillations, thereby improving the control precision of the tension closed-loop system. After system optimization, tension-thickness coupled fluctuations during thin-gauge strip rolling are notably mitigated, achieving a steady-state tension control precision of ±3% and a thickness control precision of ±2 μm.

Currently, the production of 690 MPa grade marine platform steel plates usually adopts offline heat treatment, which not only wastes a large amount of heat but also takes a long time. This paper improves the existing heat treatment process and adopts the full-process online heat treatment (i.e., online quenching + online tempering) process. The microstructure of the plates after online heat treatment is observed by OM, SEM, TEM, and EBSD, and the yield strength, tensile strength, and impact energy are tested by a universal testing machine to explore the laws of microstructure evolution and property changes of the steel plates under different online quenching temperatures. The results show that the microstructure of the experimental plates after heat treatment is mainly composed of tempered martensite, ferrite, and carbides. With the increase of quenching temperature, the amount of martensite and the width of laths in the microstructure increase, the amount of carbide precipitation increases, the proportion of high-angle grain boundaries decreases, the yield strength and tensile strength increase, and the impact energy decreases. After online quenching at 850 ℃, although the yield strength of the experimental plates is lower than that at 900 ℃, the yield-to-tensile ratio and impact toughness are higher than those at 900 ℃, and the comprehensive mechanical properties are the best. The research results provide theoery support for the technology optimization of E690 plate.

The strip crown plays a crucial role in determining the quality of products in strip hot rolling. Hence, achieving precisely hot-rolled strip crown diagnosis is important to improve the control capability of hot rolling. Since the hot rolling process features nonlinearity, heredity, and strong coupling, the diagnosis of strip crown is an imbalanced problem with complex decision boundaries. Existing crown prediction models tend to learn more information from the majority class, but ignore the data of the strip with unqualified crown. To overcome this limitation, a hot-rolled strip crown diagnosis model based on the fusion of hybrid resampling and cost-sensitive is proposed, and the cost-sensitive factor is obtained by artificial hummingbird algorithm. Some advanced machine learning models are selected as comparison models. The experimental results demonstrate that the proposed model outperforms all comparison models with the A_UC of 0.889 and defect recall of 0.870. Moreover, the testing time of the proposed model is only 0.0059 s. After the madel was applied to online diagnosis,the detection rate of defect crown strips increased from 82% to 88%,and the crown compliance rate of strips improved from 59% to 71%.

Due to harsh production environment and numerous coupling factors of control system, the process of heating furnace is always a weak link in the automation degree of production line, which affects the intelligent and digital process of production line. With the continuous update and development of big data technology, it has gradually entered the field of metallurgical industry, aiming at using data mining technology to establish data-driven mathematical models, so as to break through the bottleneck of traditional models. This paper introduces the construction of intelligent control system of heating furnace combustion based on big data platform in a certain scene. Big data algorithm is adopted to carry out intelligent optimization control for heating furnace combustion process, improve the control precision of heating furnace. The intelligent furnace temperature control boasts an online rate not less than 93% and discharge slab temperature control accuracy (±12 ℃) of 93.2%, while reducing fuel consumption of 3.5%,achieving energy savings and improve product quality.

The effective surface defects detection of strips is of great importance for ensuring product quality. However, due to low contrast and small target size, existing detection methods often face challenges in achieving sufficient detection accuracy. Therefore, a strip surface defect detection algorithm based on semantic enhancement and local attention mechanism (Scale Aware and Local Attention Detection, SALADet) is proposed. Firstly, a semantic interaction enhancement module is embedded in the backbone network to extract and enhance high-level semantic information in deep feature maps, improving the network's ability to distinguish between background and defects. Secondly, a local attention pyramid is introduced in the neck structure of the network to enhance the feature extraction of small targets, thereby improving the detection accuracy of small-scale objects. To further enhance detection performance, SALADet algorithm employs a decoupled detection head, effectively alleviating the conflict between classification and regression tasks, thus improving overall detection accuracy. Experimental results on the NEU-DET dataset show that the mean average precision of SALADet algorithm reaches 79.4%, representing improvements of 4.7%, 14.1%, 4.5%, 4.6%, and 6.1% over Faster R-CNN, SSD, YOLOX, YOLOv8 and CenterNet algorithms, respectively. Additionally, SALADet algorithm achieves an inference speed of 84.7 frames per second, demonstrating excellent real-time performance and practicality.

To solve the problem of thick oxide scale and color variation between edges and center of S355GD-M strip, in-situ observations were conducted on oxide scale formation during the heating, and experimental studies were performed on the structural transformation of surface oxide scale after strip cooling at different coiling temperatures. Results indicate that oxide scale on S355GD-M slab surfaces during heating progresses through several stages: intergranular growth, dense growth, loose growth, aggregated longitudinal growth, and stable growth. At the beginning of oxidation, the specimen surface is not covered by oxidation products, allowing Fe atoms in the substrate to fully contact O₂. Due to higher energy at grain boundaries within the substrate, oxidation products nucleate preferentially at grain boundaries. As temperature further increases, grain boundaries become fully covered by oxidation products, and the oxidation reaction proceeds toward non-grain boundary regions until the specimen surface is completely covered by dense oxidation products. At this time, the outermost layer of oxidation products in contact with O₂ exhibits the highest oxygen concentration, where FeO is oxidized to Fe₂O₃ and Fe₃O₄ and begins longitudinal growth. As the coiling temperature decreases, the proportion of eutectoid structure first increases and then gradually decreases. Conversely, as the cooling rate decreases, the proportion of eutectoid structure gradually increases. Based on experimental results, a CCT curve for the FeO structural transformation within the iron oxide scale was plotted, revealing compliance with the C-curve pattern, with a nose temperature around 500 ℃. Based on in-situ observation results, higher temperatures accelerate iron oxide scale growth. An effective strategy to reduce iron oxide scale thickness is to minimize the strip's prolonged exposure to high temperatures. By lowering the starting rolling temperature, increasing the finishing rolling temperature, and raising the strip rolling speed, the high-temperature dwell time was effectively reduced, achieving the goal of thinning the oxide scale. The structural differences in oxide scale primarily result from varying cooling rates across different strip locations. Based on experimental results of FeO eutectic phase transformation, the coiling system was adjusted by setting the coiling temperature near the nose tip temperature, thereby reducing structural variations between the strip edges and center. Following process optimization, the thickness of the oxide scale on the surface of S355GD-M strip reached about 12 μm. Both the edge and center regions exhibited an oxide scale structure composed of Fe₃O₄ + FeO + eutectic phase. This significantly improved surface color defects, enhancing the surface quality of the hot-rolled product.

Hot-rolled coils produced by the ESP production line exhibit yellowing and rusting after pickling, with significant variations in the degree of yellowing among coils subjected to different coiling temperatures. Following adjustments to the pickling process, the strip surface becomes corroded, forming brown or yellow "flowing rust". This paper therefore conducts an analysis and investigation into this issue. The results indicate that the yellowing phenomenon arises from the combined action of the condition of the iron oxide scale on the surface of hot-rolled strips and the pickling process, where the key influencing factors include the thickness of the surface iron oxide scale, pickling speed, and squeezing efficiency. Mechanistic analysis of yellowing reveals that residual hydrochloric acid on the strip surface post-pickling reacts with matrix iron to form ferrous chloride, which further oxidizes to generate iron hydroxide and hydroxyl oxides, thereby causing yellowing. Maintaining the wettability of the strip surface after pickling can prevent oxidation reactions between the strip's matrix iron and air, ultimately eliminating surface yellowing and enhancing the overall product quality.

Gearbox transmission systems typically operate at high speeds, rendering their key components susceptible to significant wear and impact damage. Given the inherent complexity of gearbox systems, simulation models often fail to accurately diagnose gearbox faults. To address this issue, this paper proposes a digital twin-based gearbox fault diagnosis method that combines physical modeling and data-driven approaches. First, by constructing two Boltzmann machines, features are extracted from the sensor data and the simulation data generated by the rigid-flexible coupled dynamic model. The information from these two data sources is mapped to a high-dimensional space to form a joint representation, which is then integrated with a multi-layer feed-forward neural network to create a multimodal information fusion model for real-time fault detection. However, during the practical application of digital twin-driven fault diagnosis methods, a discrepancy between the virtual space and the physical space often exists. Therefore, this paper develops an adaptive correction framework based on the multi-objective locust optimization algorithm to enhance the fidelity of the virtual space. Experimental results demonstrate that the proposed digital twin method reduces the information error between the physical space and the virtual space and significantly improves the accuracy of fault diagnosis of gearbox.

This paper presents the latest advancements in the production of SFLH (Steel Foil Lacerable by Hand,SFLH) and BMF (Blown Metal Foil,BMF) using rolling techniques. It defines the thickness ranges for various thin strips, super-thin strips, ultra-thin strips and foils. The process characteristics of rolling SFLH and BMF under conditions approaching the theoretical MRG (Minimum Rolling Gauge,MRG) are discussed. Several formulas for calculating MRG are introduced and compared, including the Stone formula, the LTAR formula which accounts for speed ratio differences, and the Xiao-Hong formula based on Fleck′s ultra-thin strip rolling theory. Approximate estimation formulas and curves illustrating the relationship between work roll diameter and MRG under both synchronous and asynchronous rolling conditions are provided, along with roll diameter, thickness, and speed ratio diagrams. Related theoretical issues such as the suitable rolling thickness of sheet and strip materials,and the mechanical model of negative-gap rolling are also discussed. Typical applications of SFLH and BMF in copper foil, aluminum foil, stainless steel foil, as well as in the electronics and semiconductor industries, new energy sectors, and precision structural component micro-manufacturing are introduced. Finally, future development directions for rolled BMF are outlined.

The emulsion is used as lubricant in the cold rolling of silicon steel, but it often forms emulsion spots on the surface of strip, which affects the surface quality. In this work, SEM, EDS, XPS, XRD and AFM were used to study the formation mechanism of emulsion spots on the surface of silicon steel strip. The results showed that the strip was oxidized to form the line-shape Fe₃O₄ products, which covered the surface of the strip and made the surface rough, resulting in residue of the emulsion. The water in the emulsion evaporated, while the oil in the emulsion remained in the oxidation area to form black emulsion spots. The composition of the emulsion spot on the surface of silicon steel strip is Fe₃O₄ product and oil pollution. According to the formation mechanism of emulsion spot, the formation of emulsion spots can be effectively reduced by controlling the shape of strip and strengthening emulsion management.

In the plate production process, accurate contour measurement is not only the key cornerstone for optimizing the rolling process and adjusting equipment parameters, but also a decisive factor in ensuring product quality accurately meets customer needs. Considering that the current advanced plate surface defect detection system can efficiently produce detailed image data, this paper attempt to explore a new path to integrate surface defect detection and plate contour measurement, striving to achieve the high-precision goal of contour measurement at a lower cost. By fully utilizing the high-resolution plate images output by the surface defect detection system, a series of data processing steps are implemented, including image distortion correction, precise adjustment of attitude and tangential angles, and thickness and jitter compensation algorithms based on binocular vision technology, in order to accurately restore the actual size information of the plate. During the process, a dual-position detection and calibration mechanism for the upper and lower surfaces of plate is also introduced, effectively eliminating measurement errors caused by lateral displacement during plate movement, ensuring the authenticity and accuracy of contour data. The generated contour data can not only be used for traditional length and width measurements, but also further cover advanced analytical dimensions such as rectangle evaluation and sickle bending detection, providing comprehensive and high-precision data monitoring and feedback for plate production.

In order to solve the problem that the current rolling force prediction model can not accurately reflect the actual rolling force, a new rolling force prediction model based on RBF neural network was established by revising the Sims model twice. Based on the actual production data of Q345 steel, the data set required for RBF neural network prediction is formed through the reverse calculation of the actual deformation resistance and the data normalization processing. Then the network structure is optimized to achieve the accurate prediction of deformation resistance, and the theoretical rolling force is obtained by using the modified deformation resistance calculation. In order to further improve the rolling force accuracy, starting from the Sims model structure, RBF neural network was used to modify the model proportion coefficient (the ratio of actual rolling force to theoretical rolling force), and the rolling force correction coefficient was obtained. According to the performance evaluation, the maximum error of the rolling force after first correction is 21.595%, and the maximum error of the rolling force after second correction is only 3.631%. This paper provides a new idea for improving the precision of rolling force.

During the hot rolling process, the precise positioning of steel billets is of paramount importance, as it enables automated billet inspection, which in turn enhances both production efficiency and quality control. However, the actual production process is arduous due to the harsh conditions of the billet production environment, which presents challenges for plate inspection. To address the problems mentioned above, a robust Yolov8 algorithm for billet detection is proposed. A hybrid attention mechanism network is proposed to solve the problem of shallow feature loss in pyramid networks, enhancing the model's ability to learn local features and improving detection accuracy while maintaining model lightweightness. The implementation method involves the introduction of attention modules into the network with the objective of enhancing the preservation of detailed image feature information, thereby improving the overall detection accuracy of the targets. Subsequently, the NPANet feature fusion structure is designed to enhance the network's ability to fuse multi-scale features of images. This is achieved by refining the convolution modules to make the network model lighter. Finally, the loss function is improved to enhance the algorithm's regression performance and reduce the error in stable box generation. The experimental results demonstrate that the improved NDS-yolov8 model, in comparison to the initial network structure, exhibits a reduction in the weight file size from 6.2 MB to 4.6 MB, a decrease infloating-point performance from 8.1 GFLOPS to 6.4 GFLOPS, and an increase in the mean average precision (P_mA@[0.5:0.95]by 0.5% at different I_oU values. Compared to actual values in real-world scenarios, the NDS-yolov8 network model demonstrates significantly reduced error margins relative to the original Yolov8 network model. It achieves more accurate estimation of the billet's real-time position, thereby effectively enhancing the performance of billet detection and localization.

Aiming at the problem of fatigue failure of rolls caused by alternating contact stress of Sendzimir mill, the contact stress distribution and fatigue failure mechanism of rolls were explored. Based on the influence function method, the contact pressure calculation model between rolls was established. The crack initiation life model was established by using the local stress and strain method of damage mechanics. The crack propagation life model was established by using the crack growth rate method of fracture mechanics. The real-time production data was integrated to develop the roll fatigue life prediction model system. According to the internal fatigue distribution of the roll, the grinding amount can be reasonably formulated, the over-limit fatigue layer can be removed, the safety of the roll can be guaranteed, and the rolling kilometers can be effectively increased. It is of great significance to realize precise grinding roll, precise preparation roll and precise use roll.

Aiming at the lightweight requirements for the vehicle, this paper investigates the influencing factors of the anti-concave property of automotive sheets. By combining anti-denting experiments with finite element simulations, the effects of pre-strain, sheet thickness, part curvature on the anti-concave performance of automotive sheets were studied. The experimental results show that when the pre-strain is less than 15%, the anti-concave stiffness of the sheet increases with the increase of pre-strain. When the pre-strain is greater than 15%, the anti-concave stiffness of the sheet decreases with the increase of pre-strain, which is due to the thinning of the sheet thickness caused by excessive pre-strain. For sheets with thicknesses of 0.65 mm and 0.70 mm, when the pre-strain is less than 3%, there is not much difference in the anti-concave stiffness of the sheets and the indentation depth is greater than 2 mm. When the pre-strain is greater than 3%, the anti-concave stiffness of the 0.7 mm thick sheet is significantly greater than that of the 0.65 mm thick sheet, and the indentation depth of both decreases sharply. As the pre-strain further increases, the indentation depth does not change significantly. This is because with the increase of pre-strain, the sheet undergoes work hardening, and the strength is significantly improved. The thickness of the sheet is one of the important factors determining its indentation resistance. Bake hardening has little effect on the anti-concave stiffness of sheet, but the depth of dents is significantly reduced compared to before baking. This is due to the increase insheet strength caused by strain aging during the baking process. The finite element simulation further verified that the curvature of the part is also one of the key influencing factors on its anti-concave property. The anti-concave property of the part increases with the decrease of its curvature. However,a smaller the curvature is not always better. It should be controlled within a reasonable range.

During the production process of continuous hot-dip galvanized strip, bright spot defects on the coating surface seriously affect the corrosion resistance of products. In this paper, multi-scale characterization techniques (SEM-EDS and AFM) are used to analyze the morphology and composition of bright spot defects, revealing that their main causes are zinc dross inclusions, gas pore aggregation and abnormal local alloying. An online detection method for bright spot defects based on machine vision is proposed, and an improved U-Net network is adopted to realize real-time classification of bright spot defects, finally achieving a classification accuracy of 98.2% and above, which meets the requirements of real-time performance and high precision for bright spot defect detection in high-speed production lines of continuous hot-dip galvanized strip. The correlation between coating thickness control precision and bright spot defect suppression is studied, a fuzzy PID controller integrated with dynamic weight adjustment is designed, and the feedback data of laser thickness gauge is incorporated to realize adaptive control of coating thickness (control error within ±1.5 μm). Experiments show that this scheme reduces the defect occurrence rate by 63% and increases the coating thickness qualification rate to 99.1%.

In response to the severe uneven wear of roll during the production of silicon steel using conventional parabolic roll shapes on a certain hot-rolled coil production line, local high points were prone to occur on the rolled plate for the later stage of rolling, which can cause ridge buckle for the latter process. And crown was controlled with difficulty. The article developed a set of composite roll profile curves for work rolls, which combined parabolic roll profile and double taper roll profile to improve uneven wear of rolls and enhance the quality of rolled section. The numerical simulation technique was used to establish an elastic-plastic finite element model for the composite profile of hot rolling work rolls. The simulation results showed that compared with conventional profiles, the profile have greater advantages in the efficiency of bending force control and the improvement of hot rolling profile and edge drop control level. The roll type was applied to the production site for debugging and verification. The results showed that after the new roll type was put into use, the problem of uneven wear of the roll was significantly improved. The number of silicon steel rolling blocks per unit roll period increased from 40 to 50, and the compliance rate of hot rolling profile C₄₀ ≤ 30 μm increased from 46.2% to over 90%.

Annealing is typically employed to mitigate residual stresses in materials. However, for low-carbon micro-alloyed high-strength steels, their low carbon content results in insufficient precipitation-induced plasticity strain to adequately relieve residual stresses. This study investigates the influence of plasticity behavior on residual stress at different annealing stages through stress measurement, microstructure characterization, and partitioning of plasticity strain. The results indicate that the partitioning plasticity induced by Mn partitioning behavior is greater than the precipitation plasticity caused by the precipitation of alloy carbides, making it the primary mechanism for residual stress relaxation in low-carbon steel. The partitioning plasticity strain parameter K associated with Mn partitioning is 2.691×10^-5 MPa^-1, while the precipitation plastic strain parameter K for alloy carbide precipitation is 1.303×10^-5 MPa^-1. After holding at the Mn partitioning stage for 15 minutes, the absolute residual stress of the tested steel decreases from 980 MPa to 330 MPa, a reduction of 66.3%. In contrast, after holding at the alloy carbide precipitation stage for 15 minutes, the absolute residual stress decreases from 323 MPa to 277 MPa, a reduction of only 14.2%. Therefore, partitioning plasticity is the dominant mechanism for residual stress relaxation in low-carbon steel.

Flatness control technology for cold-rolled strip is one of the core control technologies in the iron and steel industry. Its control effectiveness directly determines the post-rolling flatness quality and influences subsequent downstream processes. Flatness actuators are the key technological means for achieving flatness control. Lower accuracy in the actuator models can impair the online correction capability for flatness deviations and easily lead to flatness defects. This paper comprehensively considers the influence patterns of key rolling parameters, such as rolling force and strip geometric dimensions, on strip flatness, as well as the coupling mechanisms among different flatness actuators. Based on this, a roll bending-roll tilting coordinated setup model incorporating quadratic influence coefficients is established. The concept of an "overspray coefficient" is introduced to construct a segmented cooling flow control model that accounts for the partial flow overlap characteristics between adjacent emulsion nozzles. To enhance the accuracy and computational efficiency of the setup model, an optimization strategy for setting the coefficients in the setup model is developed based on heuristic search theory. Industrial trials are conducted using a 1 450 mm UCM five-stand six-high tandem cold rolling mill as the application platform. The results indicate that the flatness actuator setup model and strategy developed in this paper can efficiently and accurately correct online flatness deviations. They reduce the setup complexity of the model and improve the hit rate of flatness root mean square error by approximately 8%.

With the rapid development of deep-sea oil and gas engineering and other fields, the requirements for material properties are getting higher and higher. In order to improve the application performance of S32707 hyper duplex stainless steel in deep-sea oil and gas engineering, this paper studies the effect of solution temperature on its mechanical properties and corrosion resistance. According to the calculation results of Themo-Calc thermodynamic simulation software, the S32707 samples after rolling were solution treated at different temperatures (1 055 ℃, 1 095 ℃) for 1 h. The microstructure of S32707 hyper duplex stainless steel at different solution temperatures was characterized by metallographic, scanning electron microscopy, electron probe and backscattered electron diffraction. The mechanical properties and corrosion resistance of S32707 samples were measured by tensile testing machine, impact testing machine, material surface performance comprehensive tester and electrochemical comprehensive tester. When the solution temperature is above 1 055 ℃, there are only ferrite and austenite in the S32707 sample. With the increase of solution temperature, the concent of α/γ ratio increases, the content of Cr, Mo and Si in the ferrite phase decreases, the content of Ni increases, and the austenite is the opposite, that is, the content of alloying elements between the two phases gradually tends to be consistent. At higher solution temperature, the increase of ferrite phase content and the transfer of alloying elements improve the strength, hardness, and wear resistance of the samples, and reduce its plasticity, toughness and elongation. At lower solution temperature, the impedance arc of S32707 sample is larger, the anodic dissolution rate is lower, and the passivation zone is wider, indicating that its corrosion resistance is better.The research results provide data support for the production of hyper duplex stainless steel and its application in deep-sea oil and gas engineering,seawater desalination and other fields.

In response to the preparation requirements for non-oriented silicon steel used in new energy vehicle drive motors, this paper obtained normalized sheets with similar microstructures but different thicknesses by adjusting the normalization process. Subsequently, through cold rolling and annealing, non-oriented silicon steel sheets with a thickness of 0.25 mm were obtained. The effects of cold rolling reduction and annealing temperature on the microstructure, texture, magnetic properties, and mechanical properties of the non-oriented silicon steel sheets were investigated. The results showed that as the thickness of the hot-rolled non-oriented silicon steel sheets decreased (the cold rolling reduction decreased), the intensity of the unfavorable textures {223}<110> and {111}<110> in the cold-rolled sheets gradually decreased, ultimately improving the texture of the annealed sheets, with the texture factor gradually increasing by 0.14 to 0.63. The iron loss P_10/400 decreased by 0.2-1.3 W/kg, the magnetic induction intensity B₅₀ increased by 0.01-0.02 T, and the yield strength only decreased by 2-11 MPa. With an increase in annealing temperature, the P_10/400 of the annealed sheets with three different initial thicknesses decreased by 0.3-0.8 W/kg, and B₅₀first increased and then decreased (reaching a maximum value at an annealing temperature of 940 ℃), while the yield strength decreased by 3-13 MPa. Therefore, by thinning the initial hot-rolled steel sheets, the cold rolling reduction can be reduced, and combined with the optimization of the annealing process, the texture of the annealed sheets can be improved, enhancing the magnetic properties of the non-oriented silicon steel sheets.