Modeling deformation resistance for hot rolling based on generalized additive model

A model of deformation resistance during hot strip rolling was established based on generalized additive model.Firstly,a data modeling method based on generalized additive model was given.It included the selection of dependent variable and independent variables of the model,the link function of dependent variable and smoothing functional form of each independent variable,estimating process of the link function and smooth functions,and the last model modification.Then,the practical modeling test was carried out based on a large amount of hot rolling process data.An integrated variable was proposed to reflect the effects of different chemical compositions such as carbon,silicon,manganese,nickel,chromium,niobium,etc.The integrated chemical composition,strain,strain rate and rolling temperature were selected as independent variables and the cubic spline as the smooth function for them.The modeling process of deformation resistance was realized by SAS software,and the influence curves of the independent variables on deformation resistance were obtained by local scoring algorithm.Some interesting phenomena were found,for example,there is a critical value of strain rate,and the deformation resistance increases before this value and then decreases.The results confirm that the new model has higher prediction accuracy than traditional ones and is suitable for carbon steel,microalloyed steel,alloyed steel and other steel grades.     

轧制变形抗力数学模型的发展与研究动态

建立变形抗力数学模型对变形抗力预测和电子计算机在线控制生产的普及具有重要意义。从传统解析法和人工神经网络法两个角度,系统介绍了国内外建立变形抗力模型的发展应用与研究动态,结合建模中存在的问题,针对性地提出了一些建议。     

热连轧变形抗力模型的优化改进

改进后的志田茂变形抗力模型是比较经典的变形抗力模型,但该模型没有考虑残余应变对于变形抗力的影响,不能完全适应现场生产的需要。作者在分析研究济南钢铁集团有限公司热连轧厂大量实际生产数据的基础上,在原变形抗力模型中考虑了碳质量分数的影响,增加了对应的影响系数,用于修正残余应变的影响,从而使该模型更贴近于现场实际。模型实际应用结果表明,改进后的模型预报精度明显提高,效果良好。     

热轧相变过程变形抗力模型研究与开发

 对精轧阶段存在相变的热轧钢种,因变形抗力随轧制温度的变化规律与常规的奥氏体轧制钢种显著不同,使得传统变形抗力模型的预报误差较大,严重影响这类钢种的轧制稳定性。为此,研发了一种热轧相变过程变形抗力模型,通过在原变形抗力模型基础上添加一个新的相变趋势项,该修正项为轧制温度的二次多项式函数,并根据钢种分类来精细优化适应不同钢种轧制的多项式待定参数。该模型目前已成功应用于涟钢CSP热连轧生产线变形抗力在线计算,实际生产应用表明,新模型上线后,变形抗力与轧制力的预报精度显著提高,轧制力模型预报误差12%以内的比例从83.3%提高到96.7%,满足了热连轧精轧相变带钢的稳定生产要求。     

热轧带钢组织性能预报模型及应用

 基于物理冶金理论,研究了热轧带钢过程中的奥氏体晶粒长大模型、奥氏体再结晶模型、奥氏体相变模型以及力学性能模型。奥氏体再结晶模型中,通过研究位错密度的变化来描述由于再结晶不完全造成的变形抗力的变化。奥氏体相变模型中,通过碳扩散理论描述了奥氏体-铁素体相界面随冷却过程的变化规律。基于热轧带钢过程中的冶金物理模型,开发热轧带钢组织性能预报系统。系统包括4个模块,分别用于计算板坯在加热炉、粗轧精轧、层流冷却和卷取完成各阶段的组织和力学性能参数,生产工艺是该系统的重要输入参数。利用该系统对某钢厂实际生产过程的组织性能进行预报,预报的力学性能和现场实测值有较好的一致性。     

基于增量分析和PSO-LSSVM的热轧辊缝调平预测模型

辊缝调平在保证热轧带钢板形质量和轧制稳定性中起着关键作用,目前以操作人员目测后经验调整为主,无法满足未来少人化、智能化轧制技术需求.为此,基于增量分析方法实现工艺参数增量因子提取,有效解决传统离散数据预测中部分信息丢失问题;同时以粒子群算法(PSO)优化最小二乘支持向量机(LSSVM)模型参数,使得参数选取更加科学.采...     

基于轧制过程数据的轧机刚度动态模型

针对实际轧制过程中轧机刚度变化难以在线测试的难题,对轧制过程进行数据挖掘,提出了基于厚差溯源分析的轧机弹跳和轧机刚度定向挖掘方法。对某650可逆冷轧机的轧制过程数据进行了分析,并深入挖掘了轧机刚度和轧件宽度、轧制速度的关系,建立了轧机刚度的综合动态模型,对提高板厚控制精度具有重要的意义。     

一种建筑用耐火钢变形抗力模型的建立

利用Gleeble-1500热模拟试验机对一种建筑用耐火钢进行了压缩试验，分析了变形温度、变形速率以及变形量对变形抗力的影响；在变形抗力σ与温度T的关系中，考虑了应变ε和应变速率ε的影响；在应变速率ε的影响指数中，考虑了温度的影响，建立了一种含Nb、Ti等微合金元素建筑用耐火钢应力峰值前的变形抗力模型。     

热轧前滑模型的研究

前滑模型对热轧生产参数的设定有重要意义。基于Orowan微分方程和Sims假设,建立了热轧前滑理论模型,并把该理论模型中相关的3个参数(轧辊半径、入口厚度、出口厚度)优化为2个无单位的参数,采用统计分析法得到简易的经验模型,并分析了各因素的影响。现场实测数据表明:前滑模型计算的最大绝对误差是0.001,最大相对误差是6.93%,能满足在线生产要求。     

冷连轧变形抗力模型的优化

采用原始静态变形抗力模型,并考虑入口、出口弹性变形轧制力对模型的道次影响系数,进行计算优化,建立了1个优化的变形抗力模型。2种模型的对比表明,优化的模型精度更高,提高了轧制力的计算精度。     

A brief investigation on transversal temperature distribution during hot rolling

Temperature evolution of a working piece during hot rolling has a significant influence on the microstructure and final mechanical properties of the material. As the measurement technologies improve,on-line measurement of the temperature across the strip width has become accessible,and thus monitoring and control of the transversal temperature distribution during hot rolling can be realized.The scanning of the temperature across the strip width can help us to understand the state of the temperature variation at the strip edges and as a result help us improve the temperature homogeneity of the strip.In this paper,reaearch on temperature distribution along the strip width is reviewed first,and then the mechanism of the scanning measuring devices is introduced.With the temperature scanning measurement data taken from the finishing mill entry point and the down coiler entry point,the temperature distribution at the strip edges is analyzed.It is pointed out that the roughing process is the main contributor to the temperature inhomogeneity.Furthermore,the contribution of the edger heater to the temperature homogeneity is investigated.     

Influence of process variables on development of camber during hot rolling of strip steel

Abstract

Experimental design techniques have been used to study the effects of variations in horizontal scale breaker process parameters in relation to camber generation during the early reduction stages in a hot strip mill. A fractional factorial experiment was designed and performed to identify which process variables are most influential to camber generation. The parameters which emerge from this analysis as being significant are: slab wedge, temperature differentials across the strip width, the deviation of the slab from the mill centreline, and mill tilt. A full factorial experiment was then designed and performed to further investigate these significant process parameters to determine where to set influential and controllable variables so that camber is minimised. The results show that none of the process parameters is significant to camber generation when acting alone; however, they are all significant when combined with one or two other process parameters. This study has enabled a predictive equation to be developed which is able to predict camber to within ±4 mm. This equation may be used in conjunction with mill tilting strategies to reduce camber in the early reduction stages of a hot strip mill.     

热轧带钢轧制过程的稳定性

在分析热轧带钢时产生带钢运行故障原因的基础上，对带钢粗轧和精轧等过程提出了相应的控制措施．提高了带钢轧制过程的稳定性，保证了带钢产品的质量．     

基于RF-LSTM的热连轧板坯粗轧出口温度预报

热连轧板坯温度的连续、实时预报是提高带钢产品性能的关键,针对热连轧过程数据维度过高不利于快速、准确预测板坯温度的问题,提出一种基于随机森林-长短期记忆神经网络(RF-LSTM)的板坯粗轧出口温度预测模型.首先,采用改进随机森林算法对特征变量进行选择,通过分析板坯粗轧出口温度的预测结果变化衡量特征的贡献度,进而构造反映过...     

一种新型实用热轧工艺参数模型的开发

介绍了一套自主开发的热轧工艺参数模型．该模型内耦合了不同钢种的变形抗力曲线,这些变形抗力方程中耦合了钢的化学成分、温度、应变、应变速率及奥氏体晶粒尺寸等因素．根据输入的工艺参数用西姆斯方程计算每道次的应变速率及应变量,并得到相应道次的变形抗力、热轧轧制力、力矩及功率等参数．模型可根据实测的结果自学习,并修正相应的结果．与攀钢热轧厂的实测结果相比,模型的输出结果吻合较好,预测误差在10%以内．     

Mechanical property prediction model of hot- rolled strip via big data and metallurgical mechanism analysis

A global additive model for mechanical property prediction of hot- rolled strip was proposed, which combined metallurgical mechanism and industrial big data.First, a set of systematic and general modeling methods were presented, including: comprehensive use of the data mining methods such as random forest, Bayesian network, causality diagram, and the metallurgical mechanism and priori knowledge were combined to determine the influence factors of the model. The construction steps of unitary generalized additive model without interactions were presented, which included data cleaning, sub- model building, sub- model validation, sub- model modification.Then the interactions between the various factors were explored, and the binary generalized additive models with interactions were constructed; then, the relationships between the sub- models were analyzed, and these sub- models were combined to construct the whole model. Last, the application of the above modeling methods were verified, a prediction model of mechanical property of micro- alloyed steel products was established based on a large number of production data of hot- rolled Nb micro- alloyed steel.The effects of composition and process parameters on the mechanical properties of steel are obtained, and the prediction error of the new model for tensile strength is 2. 51%. The above research provides a new modeling idea for complex industrial systems.     

中厚板轧制过程控制中轧件的平均温度模型

通过分析得出,在中厚板轧制过程中轧件厚度方向上的温度分布是以轧件中心为原点的抛物线形式;考虑到表层变形区和中心变形区对轧制变形的影响作用的不同,建立了与压下率、表面温度和中心温度有关的平均温度模型;将有限差分法处理实测温度后获取的表面温度和中心温度应用于此模型,并将模型应用于国内某中板厂3 000 mm轧机过程控制的轧制力预测中,获得了良好的效果。     

热轧带钢轧制力模型自学习算法优化

根据轧制数量、测量数据质量和轧制力预报误差的影响,建立了轧制力自学习速度因子优化模型.在长期自学习的判定条件中综合考虑了规格分档的变化以及厚度、宽度的改变量,减少了换规格的次数.长期自学习系数计算时利用了从前一块钢数据中分离出的设备状态信息,从而改善了模型自学习的连续性.离线仿真分析结果表明,轧制力计算精度相对于自学习算法优化前有较大的提高.     

基于热轧全流程氧化铁皮控制的耐蚀性工艺

针对热轧生产流程实际工况,系统研究了热轧、卷取阶段的三次氧化铁皮演变规律,旨在不增加生产成本的前提下,通过调整热轧生产工艺控制氧化铁皮结构,利用热轧后生成的氧化铁皮作为防护屏障,提高钢材耐蚀性能.结果表明,在不同轧制温度下,三次氧化铁皮结构从外到里分别为Fe2O3、Fe3O4和FeO,由于FeO中的阳离子空位密度大,导...     

热连轧精轧机组温度控制数学模型研究

从现场实际出发研究了精轧各个环节的温度变化过程,分析了带钢热连轧终轧温度与各影响因素之间的关系.同时将温度模型程序化,采用计算机系统进行分析和计算,提出了行之有效的温度模型方案.该温度模型对其它热轧生产线同样适用.