This study designed and fabricated three low-magnetic stainless steels Fe67-xCrxNi25Mn1Mo2Ti4Al1 (X = 13, 18, 25, at.%, designated as 15A3-1, 15A3-5 and 15A3-6), investigating the influence of chromium content on magnetic and mechanical properties. The 15A3-1 steel exhibits superior strength-ductility synergy, with post-aging properties including a yield strength of 722 MPa, tensile strength of 1185 MPa, and elongation of 26.1%. The primary precipitates consist of discontinuous lamellar gamma ' and continuous spherical gamma ' precipitates, both possessing an identical L12 crystal structure while maintaining coherency with the matrix. In 15A3-5 steel, brittle Heusler phases precipitated near grain boundaries reduce elongation. The 15A3-6 steel develops 1.7 vol% martensite after solution treatment and 12.7 vol% martensite following aging treatment. Reverse martensitic transformation refines the post-solution grain size to 19.5 mu m, while aging induces multiphase precipitation of gamma ', Heusler, and sigma phase. The B-H curves of 15A3-1 and 15A3-5 exhibit linear responses, confirming paramagnetic behavior. Conversely, 15A3-6 displays a steep initial slope in its B-H curve at low applied fields; after saturation magnetization of its ferromagnetic phases, magnetic induction increases linearly with applied field. Measured relative permeabilities are 1.00326, 1.00457 and 9.33065 for 15A3-1, 15A3-5 and 15A3-6 respectively. Dilatometric measurements indicate that the Ms temperatures of 15A3-1 and 15A3-5 are below 25 degrees C, while 15A3-6 has an Ms temperature of 30 degrees C. Coherent precipitates generate localized stress fields that provide preferential nucleation sites for martensite. Cr enrichment and Ni depletion in the matrix adjacent to Ni3(Al,Ti) further elevate the Ms temperature. Although stacking fault energy (SFE) decreases with increasing Cr content, all steels maintain SFE >40 mJ m-2, resulting in dislocation slip as the dominant deformation mechanism.

The classical Bland-Ford model serves as the foundation for calculating the rolling force in cold rolling process, derived based on six assumptions and approximate conditions. This study eliminates the shear stress assumption ( p_x - σ _x=2k ) and the differential term assumption ( (p_x/2k - 1)d/dφ(2h_xk)<<2h_xkd/dφ(p_x/2k) ) from the classical model. The Bland-Ford cold rolling force model with modification of shear stress and differential term is derived. The problems of the underestimation of the neutral angle and the overestimation of the normal pressure at the neutral angle for the classical Bland-Ford model are effectively solved. Based on measured data from a 1340 mm cold rolling mill, a comparative analysis was conducted on the normal pressure calculation results of the model before and after modification under different specifications and friction coefficients. The computational results of applying the modified Bland-Ford cold rolling force model to 1506 sets of measured data indicate that the MSE decreased from 0.391 MN to 0.291 MN, the RMSE decreased from 0.625 MN to 0.541 MN, the MAE decreased from 0.455 MN to 0.389 MN, and the coefficient of determination R² increased from 0.678 to 0.794. E_s υ _s α γ p_γ h_x μ φ σ _x p_x τ _xy

A thermo-mechanical treatment process was developed to grain boundaries optimization in Cr-Ni-Mn-Mo-N type low-magnetic stainless steel hot-rolled plates. The effects of combined hot rolling and solution treatment on the evolution of special grain boundaries (SGBs) in low-magnetic stainless steel were analyzed using electron backscatter diffraction (EBSD). During solution treatment of conventional hot-rolled and non-recrystallization region rolling with 66.7 % reduction ratio samples, the formation mechanism of low-Sigma coincidence site lattice (CSL) boundaries was primarily the Sigma 3 proliferation mechanism, yielding a special grain boundary ratio of 50-60 %. For the non-recrystallization region rolling with 28.5 % reduction ratio sample, the formation mechanism of low-Sigma CSL boundaries in solution treatment was primarily the new twin mechanism, achieving a special grain boundary ratio of up to 70.7 %. Tensile test results indicate that grain-optimized samples exhibit higher strength and ductility than non-optimized samples. The combined enhancement of strength and ductility in low-magnetic stainless steel after grain boundary optimization is largely attributed to the formation of a highdensity network of SGBs within the grain.

Rolling force is a critical parameter in the control of the cold tandem rolling process, significantly impacting both thickness control accuracy and flatness control accuracy. The Bland-Ford model is currently the most widely used rolling force model in cold continuous rolling. However, this model uses the principal stress yield criterion and neglects the influence of shear stress, leading to increased calculation errors as the strip thickness decreases. This study proposes using the Mises yield criterion, which considers shear stress, to modify and reconstruct the Bland-Ford normal pressure model and rolling force model. The comparison of the normal pressure values before and after modification shows that the classic Bland-Ford normal pressure model, due to its neglect of shear stress, results in overestimated normal pressure values, with the maximum deviation occurring at the neutral angle. As the rolling thickness decreases and the reduction and friction coefficient increase, the calculation deviation also increases. Based on measured data from a 1340 mm four-stand cold rolling mill, the rolling forces of the final stand for different specifications and steel grades were calculated. The results indicate that the modified Bland-Ford rolling force model yields results almost identical to the classic Bland-Ford model for thick specifications (exit thickness >1 mm). However, for thin specifications (exit thickness <= 1 mm), the overall calculation accuracy can be improved by up to 1.87 %.

Increasing the yield strength of low-magnetic stainless-steel hot-rolled plates without sacrificing ductility, corrosion resistance, and maintaining stable paramagnetism is a tough technical challenge. This study successfully prepared two types of Fe-Cr-Ni-Mn-Mo-0.37/0.47 N low magnetic stainless-steel hot-rolled plates with excellent corrosion resistance, paramagnetic properties, high yield strength (621 MPa), and good ductility (47.2 %) through nitrogen alloying. The influence of nitrogen alloying on the microstructure, mechanical properties, plastic deformation mechanisms of low-magnetic stainless steel was systematically investigated. The experimental results demonstrated that nitrogen alloying significantly increased the solid solution strengthening effect under the same microstructure condition, increasing the nitrogen content from 0.37 wt% to 0.47 wt% raised the yield strength of low magnetic stainless-steel by 146 MPa. The plastic deformation mechanism of low magnetic stainless-steel involves deformation twinning and slip. The increase of nitrogen content enhanced the stacking fault energy and short-range ordering of low magnetic stainless steel, inhibited deformation twinning during plastic deformation, and promoted the plane slip of dislocation. The high density slip bands and micro bands improved the strain hardening ability of the experimental steel. The primary reason for the difference in ductility was attributed to the varying degrees of twinning-induced plasticity observed during the late stages of plastic deformation.

Herein, disordered special quasirandom structure models of Fe 20− x − y Cr 5 Ni 7 Al x Ti y ( x , y = 1, 2, 3, 4, 5) are established for the single‐phase austenitic structure of low‐magnetic stainless steel through first‐principles calculations. The alloy volume exhibits linear expansion with increasing Al and Ti content, affected by the combined effects of doping atomic radius and interatomic interactions. The Fe 14 Cr 5 Ni 7 Al 2 Ti 4 alloy demonstrates the smallest volume and higher structural stability. Energy analysis reveals that the total energy, cohesive energy, and formation energies of all systems are negative, confirming the structural stability of the crystals with varying Al and Ti content. With increasing Ti/Al atomic ratios, the bulk modulus gradually decreases, while the shear modulus and Young's modulus increases, suggesting reduced resistance to volumetric deformation but enhanced resistance to shear and tensile/compressive deformation. Low‐magnetic stainless steels with Ti/Al ratios of 1/5, 3/3, and 5/1 are fabricated to verify the calculated results. These alloys maintain stable austenitic structures and paramagnetic behavior, with yield strength, tensile strength, and elongation ranging in 476–834 MPa, 896–1139 MPa, and 14.6–41.2%. Relative magnetic permeabilities are measured as 1.00457, 1.00474, and 1.00557. This study provides theoretical guidance for the compositional optimization and technological development of high‐strength stable austenitic Fe–Cr–Ni–Al–Ti alloys.

The roll deformation model of the six-high rolling mill is one of the core models of the strip shape control theory. The influence function method (IFM) is a numerical method applied to solve the roll deformation problem. This study aims to address the problems of slow calculation speed and insufficient calculation accuracy of IFM in calculating the roll deformation of the six-high rolling mill. Three optimization measures are proposed, namely, optimizing the iterative calculation order of the roll deformation to improve the calculation efficiency; introducing the Adam (Adaptive Moment Estimation) gradient descent optimization algorithm to enhance the stability of the iterative process; and introducing a high-precision rolling force model based on the XGBoost (eXtreme Gradient Boosting) algorithm to improve the calculation accuracy of flatness in the IFM. Parallel experimental results show that by applying the three optimization measures simultaneously, improving the iterative calculation order of the roll deformation can improve the calculation speed by about 7.64 times; introducing the Adam algorithm can reduce the oscillation range of roll contact pressure by an average of 50.4%, increasing the stability of the calculation process; and introducing a high-precision rolling force model based on the XGBoost algorithm can improve the flatness calculation accuracy by about 39.1%. This study provides an effective method to improve the calculation speed and accuracy of the roll deformation in a six-high rolling mill, which has important academic application value.

Developing low magnetic stainless-steel plates with high yield strength while ensuring good plasticity, corrosion resistance, and paramagnetism is a tough technical challenge, especially under conditions of limited rolling mill capacity and total reduction. In this study, a low magnetic stainless-steel plate with excellent yield strength (620 MPa), ultimate tensile strength (891 MPa), and ductility (42.7

A newly designed Fe – Cr – Ni – Mn – Mo – N low magnetic stainless steel was tested based on hot deformation at 950–1150 °C and 0.01 – 10 s −1 strain rate conditions. The flow curves of the hot deformation process are of dynamic recrystallization type, the softening mechanism is mainly dynamic recrystallization. Complete dynamic recrystallization occurs when the deformation temperature is higher than 1100 °C. The microstructure evolution of the dynamic recrystallization process is characterized by electric backscatter diffraction. The Avrami kinetics of low magnetic stainless steel under different deformation conditions are established to effectively predict the dynamic recrystallization behavior during hot deformation. Through microstructure analysis, it is clarified that the discontinuous dynamic recrystallization formed by the high angle grain boundaries bulging and migration is the dominant dynamic recrystallization mechanism of low magnetic stainless steel. The continuous dynamic recrystallization formed by the sub-grains rotation and dislocations absorption is the auxiliary mechanism. The characteristics of dynamic recrystallization nucleation, grain growth, and microstructure evolution of the low magnetic stainless steel during hot deformation are described.

To solve the production problem that work roll bending force was inclined to a positive limit, namely bending force saturation, during rolling with large reduction rate in four stands 6-hi HC tandem cold mill, work roll contour of the last mill could be designed as sextic polynomial curve. The work roll with positive crown could decrease the quadratic crown and quartic crown of the loaded roll gap, and enhanced the shape control capability of the rolling mill. Meanwhile, wear crown of the intermediate roll could be online compensated by the crown work roll and bending force was then decreased accordingly. The appropriate maximum crown of the work roll should be comprehensively decided according to the rolling varieties, product thickness and width specifications, and the intermediate roll period. Rolling test shows that utilizing flat-roll type work roll at inception stage of intermediate roll period and installing convex-roll type work roll in mid and late stage can realize the good crown match between intermediate roll and work roll. Hence, the work roll bending force is always in better regulation range during continuous rolling. The adaptability to rolling varieties and width specifications is significantly enhanced. The proportion of composite flatness value less than 8 IU can be increased by more than 10%.

为了持续提升材料成型及控制工程专业卓越班学生在轧制成形领域的工程实践能力和水平,教学中采用宽厚板控制轧制与控制冷却的虚拟仿真实验,能实现信息技术与实践教学的深度融合.文章通过线下实验室参观和线上仿真操作相结合的方式进行工程实践,有效地提高学生对轧制设备、轧制方法、材料研发手段的认知水平,拓展了学生的工程视野.充分利用实验室资源,引入课程思政元素,增强了学生的专业自豪感和对专业社会价值的认同感,激发了其专业学习的热情,促进了创新能力的培养.

To address issues with saturated work roll bending force (WRB), maximum contact pressure between the work roll and intermediate roll (QWI), and challenging flatness lifting in a four‐stand, six‐high tandem cold mill with a 1340 mm width, the density‐based spatial clustering of applications with noise algorithm and the non‐dominated sorting genetic algorithm II are used for optimal design and online application research of a five‐segment work roll profile. When compared to a flat roll and a parabolic roll profile with a maximum crown of 15 μm, the optimized roll profile decreases the secondary crown of the roll gap (CW2) by 5.99 and 1.055 μm, respectively, and the fourth crown of the roll gap (CW4) by 0.21 and 0.39 μm, respectively. Additionally, QWI in the field of roll gap crown adjustment decreases by 1.04 and 0.445 kN mm −1 . Industrial testing shows that the average integrated value of flatness in rolling the four extreme specifications of 1340 mm tandem cold‐rolling mills lowers by 2.65 IU following the application of the new roll profile, and the average reduction of WRB in the last stand is 2.215 MPa.

为促进小班化教学质量的稳定性和均衡性,增强学生参与感和获得感,利用监督评价和学生评价信息对材料成形自动控制基础课程教学协同状态进行内观和及时调整,利用教学目标达成评价结果对各班教学情况、学生学习和考核情况进行总结,推动课程团队教师采取更有效的措施,提升教学质量和水平.

将科研成果与教学内容融合,进行特色创新,拓宽学生视野,通过具体科研实例,培养运用知识解决实际问题能力.在知识传授的同时进行价值引领,通过课程中贯穿始终的思政教学,使学生具备社会责任感和工程职业道德、组织管理能力和终身学习能力等基本素质.

Flatness is a key quality indicator of tandem cold-rolled strip. Tandem cold-rolled production is a multi-stand simultaneous rolling process. The prediction of flatness is a typical spatial sequence data prediction problem by considering not only the complex generation mechanism but also the spatial dimension dependence. However, the currently available mechanistic and machine learning models for flatness prediction only focus on its highly nonlinear characteristics and ignore its complex spatial correlation. In addition, the over-parameterized “black box” nature of machine learning models has often been questioned. Based on this, a long-short-term memory model with an attention mechanism (Attention-LSTM) is proposed in this paper. The model is structured as a two-layer LSTM network to fully learn the highly nonlinear and complex spatial data correlation of tandem cold-rolled strip flatness. Furthermore, attention mechanisms are also added to the spatial dimension and flatness feature vector dimension, respectively, to enhance the interpretability of the model. The superiority of the proposed model is verified by comparing the error back propagation neural network model (BPNN), the extreme gradient boosting algorithm model (XGBoost), the deep neural network model (DNN), the LSTM model without the attention mechanism, and the Attention-LSTM model with the same dataset and the same optimum training method. The interpretation of the attention mechanism weights is consistent with the mechanistic model, which enhances the credibility of the model. The results of the ablative experiments also validate the effectiveness of incorporating the attention mechanism in the model.

针对攀钢冷轧厂1220 mm HC冷连轧机组板型调控能力不足的问题,设计了凸度为10 μm至80μm的8组工作辊辊型曲线,建立了带钢冷轧过程三维有限元模型,并对8组辊型曲线分别进行模拟计算.通过对计算结果进行分析,掌握了工作辊凸度对辊间压应力及带钢平直度的影响规律:8组凸度辊的辊间压应力均比平辊更小且随着工作辊凸度的增大而减小.当凸度增大至50 μm时,工作辊与中间辊辊间压应力在传动侧的峰值消失;平直度分布随着辊凸度的增大呈边浪-肋浪-肋浪+微中浪-中浪的变化规律,确定了带钢平直度峰值最小时的辊凸度为30 μm.工业试验结果表明:凸度为30 μm的工作辊辊型可以有效减小平直度峰值,增大弯辊力调控空间.

为准确反映精轧轧制参数和微观组织转变对终轧温度的影响,利用自适应线性神经网络(Adaline)和径向基神经网络(RBF)技术建立了热焓修正系数预报网络作为终轧温度长继承计算模型.首先基于热焓形式的导热偏微分方程建立了带钢终轧温度计算模型,并对带钢在辊缝变形区产生的变形功、摩擦功、与工作辊的接触导热以及机架间冷却换热进行了模型描述;然后从温度与热焓之间的转换关系入手,确定将精轧区域热焓修正系数作为终轧温度模型的自适应参数,并利用复合神经网络技术建立了由19个输入节点,20个RBF隐含层节点,20个Adaline隐含层节点和1个输出节点构成的热焓修正预报网络.结合现场数据,描述了该预报网络训练样本的构成、数据标准化处理方法,同时给出了典型的网络参数和网络的预报能力.

The problems of the strip flatness defects are always severe in the tandem cold rolling process. It is of great significance to predict flatness for flatness control according to the process conditions of products. A prediction model based on convolutional neural network (CNN) was developed in this paper, and it can accurately predict strip flatness under various conditions. According to the distribution characteristics of industrial data collected in the rolling process, the isolated forest algorithm was used to eliminate outliers. Considering the special requirements of CNN on the dimension of input features, the data folding method was used to process the input features. Additionally, since strip flatness data is a vector rather than a scalar, and the length of this vector varies with strip width, which decreases the network's training accuracy. To deal with the problems, the loss function was modified. Taking the Inception module as the basic network structure and inspired by Wide & Deep learning, a strip flatness prediction model with high accuracy was developed. The optimal architecture and parameters of our network were determined through a lot of experimental explorations. The performances of BPNN (Back Propagation Neural Network), DNN (Deep Neural Network), and the proposed model were compared by mean square error (MSE) and coefficient of determination (R2). The result indicates that the proposed model has the highest prediction accuracy and better adaptability. It has the lowest MSE, 0.9891, and the highest R2, 0.9555. Finally, the fitting coefficients of Legendre polynomials were used to further prove the excellent prediction performance of the proposed model for strip flatness. Compared with other prediction models, it can obtain the lowest prediction error for the first quadratic and quartic components of strip flatness and it can be well-applied to tandem cold rolling production.

为提高头尾板形质量,均匀带钢性能,在1 780mm线过程自动化控制的基础上开发U型冷却方式.U型冷却方式投用采用的接口方式有三类即PDI数据(ED数据库表)、冷却策略(冷却计划号)和人工干预(HMI).为实现部分钢种尾部自动投用U型冷却方式,采用人工干预的方式赋值参数,在花纹板钢种中通过对尾部U型冷却长度和U型冷却温度赋值的方式试验程序的有效性,实现了头尾冷却长度80 m头尾冷却温度高于本体80℃的功能.应用U型冷却功能有效降低了 1 780mm线尾部亮带引起的板形问题,提高了板形质量.

将"对分课堂"教学模式引入压力加工车间设计课程教学.精讲车间设计遵循的一般原理和原则,引导学生以课堂阅读专业文献、作业等形式完成自主学习、协同学习,并通过小组讨论和课堂汇报进行学习.通过点评和总结对学生学习效果进行肯定、梳理、归纳和评价.实践表明,引入"对分课堂"教学有利于学生面向工程实际开阔视野,自主运用专业知识对车间设计相关问题进行综合分析和深度思考.