Prediction of lime utilization ratio of dephosphorization in BOF steelmaking based on online sequential extreme learning machine with forgetting mechanism

The machine learning models of multiple linear regression(MLR),support vector regression(SVR),and extreme learning ma-chine(ELM)and the proposed ELM models of online sequential ELM(OS-ELM)and OS-ELM with forgetting mechanism(FOS-ELM)are applied in the prediction of the lime utilization ratio of dephosphorization in the basic oxygen furnace steelmaking process.The ELM model exhibites the best performance compared with the models of MLR and SVR.OS-ELM and FOS-ELM are applied for sequential learning and model updating.The optimal number of samples in validity term of the FOS-ELM model is determined to be 1500,with the smallest population mean absolute relative error(MARE)value of 0.058226 for the population.The variable importance analysis reveals lime weight,initial P content,and hot metal weight as the most important variables for the lime utilization ratio.The lime utilization ratio increases with the decrease in lime weight and the increases in the initial P content and hot metal weight.A prediction system based on FOS-ELM is applied in actual industrial production for one month.The hit ratios of the predicted lime utilization ratio in the error ranges of±1%,±3%,and±5%are 61.16%,90.63%,and 94.11%,respectively.The coefficient of determination,MARE,and root mean square error are 0.8670,0.06823,and 1.4265,respectively.The system exhibits desirable performance for applications in actual industrial pro-duction.

Numerical Simulation of Decarburization in a Top-Blown Basic Oxygen Furnace

An improved mathematical model to describe the decarburization process in basic oxygen furnaces for steelmaking is presented in this work. This model takes into account those factors or parameters that determine the bath-oxygen impact area, such as the cavity depth, the lance height, the number of nozzles and the nozzles diameter. In the thermal issue, the model includes the targeted carbon content and temperature. The model is numerically solved, and is validated using reported data plant. The oxygen flow rate and the lance height are varied in the numerical simulations to study their effect on the carbon content and decarburization rate.

Slag Splashing in a Basic Oxygen Furnace under Different Blowing Conditions

The influence of three different blowing conditions on the slag splashing process in a basic oxygen furnace for steelmaking is analyzed here using two-dimensional transient Computational Fluid Dynamics simulations. Four blowing conditions are considered in the computer runs: top blowing, combined blowing using just a bottom centered nozzle, combined blowing using two bottom lateral nozzles, and full combined blowing using the three top and the three bottom nozzles. Computer simulations show that full combined blowing provides greater slag splashing than conventional top blowing.

End-point dynamic control of basic oxygen furnace steelmaking based on improved unconstrained twin support vector regression

In order to improve the end-point hit rate of basic oxygen furnace steelmaking,a novel dynamic control model was proposed based on an improved twin support vector regression algorithm.The controlled objects were the end-point carbon content and temperature.The proposed control model was established by using the low carbon steel samples collected from a steel plant,which consists of two prediction models,a preprocess model,two regulation units,a controller and a basic oxygen furnace.The test results of 100 heats show that the prediction models can achieve a double hit rate of 90%within the error bound of 0.005 wt.%C and 15℃.The preprocess model was used to predict an initial end-blow oxygen volume.However,the double hit rate of the carbon con tent and temperature only achieves 65%.Then,the oxygen volume and coolant additi ons were adjusted by the regulation units to improve the hit rate.Finally,the double hit rate after the regulation is reached up to 90%.The results indicate that the proposed dynamic control model is efficient to guide the real production for low carbon steel,and the modeling method is also suitable for the applications of other steel grades.

Numerical Analysis of Slag Splashing in a Steelmaking Converter

Some variables that influence the slag splashing phenomenon in an oxygen steelmaking converter are numerically analyzed in this work. The effect of lance height, jet velocity, jet exit angle and slag viscosity on the washing and ejection mechanisms of slag splashing is studied employing transient two-dimensional computational fluid dynamics simulations. A parameter here called average slag volume fraction is proposed for the quantitative evaluation of the slag splashing efficiency. Besides, a qualitative comparison is made between the computational fluid dynamics results and physical model results from literature.

A process model for BOF process based on bath mixing degree

The process model for BOF process can be applied to predict the liquid steel composition and bath temperature during the whole steelmaking process. On the basis of the traditional three-stage decarburization theory, the concept of mixing degree was put forward, which was used to indicate the effect of oxygen jet on decarburization. Furthermore, a more practical process model for BOF steelmaking was developed by analyzing the effect of silicon, manganese, oxygen injection rate, oxygen lance height, and bath temperature on decarburization. Process verification and end-point verification for the process model have been carried out, and the verification results show that the predic- tion accuracy of carbon content reaches 82.6% （the range of carbon content at the end-point is less than 0. 1wt%） and 85.7% （the range of carbon content at end-point is 0. 1wt% -0.7wt%） when the absolute error is less than 0.02wt% and 0.05wt%, respectively.

Memetic algorithms-based neural network learning for basic oxygen furnace endpoint prediction

Based on the critical position of the endpoint quality prediction for basic oxygen furnaces （BOFs） in steelmaking, and the latest results in computational intelligence （C1）, this paper deals with the development of a novel memetic algorithm （MA） for neural network （NN） lcarnmg. Included in this is the integration of extremal optimization （EO） and Levenberg-Marquardt （LM） pradicnt search, and its application in BOF endpoint quality prediction. The fundamental analysis reveals that the proposed EO-LM algorithm may provide superior performance in generalization, computation efficiency, and avoid local minima, compared to traditional NN learning methods. Experimental results with production-scale BOF data show that the proposed method can effectively improve the NN model for BOF endpoint quality prediction.

Application of variable-filtrating technique on fuzzy-reasoning neural network system predicting BOF end-point carbon content

Artificial intelligence techniques have been used to predict basic oxygen furnace(BOF) end-points. However,the main challenge is to effectively reduce the input nodes as too many input nodes in neural network increase complexity,decrease accuracy and slow down the training speed of the network.Simply picking-up variables as input usually influence validity of model.It is quite necessary to develop an effective method to reduce the number of input nodes whereby to simplify the network and improve model performance.In this study,a variable-filtrating technique combining both metallurgical mechanism model and partial least-squares(PLS ) regression method has been proposed by taking the advantages of both of them,i.e.qualitive and quantative relationships between variables respectively.Accordingly,a fuzzy-reasoning neural network(FNN) prediction model for basic oxygen furnace(BOF) end-point carbon content based on this technique has been developed.The prediction results showed that this model can effectively improve the hit rate of end-point carbon content and increase network training speed.The successful hit rate of the model can reach up to 94.12%with about 0.02% error range.

转炉炼钢动态过程预设定模型的混合建模与预报

准确预报转炉炼钢动态过程的补吹氧气用量和冷却剂添加量,对于提高终点命中率具有重要意义·采用机理模型及基于数据的自适应神经模糊推理系统混合建模方法建立了转炉炼钢动态过程预设定模型·用减法聚类,最小二乘法及梯度下降法辨识了T S模型并用该模型对机理模型进行补偿建模·对一座180t转炉的实测数据进行了仿真,仿真结果表明该方法是切实可行并有效的·

基于火焰动态形变特征的转炉炼钢终点判定

炼钢转炉炉口火焰的动态形变特征与吹炼数据有密切的关系,在不同吹炼时期显示出不同的规律性,是吹炼过程中的显著特征。准确地表示和描述火焰边界动态形变对依据火焰图像判定转炉吹炼终点具有重要意义,且能克服静态边界特征存在的振荡剧烈问题。提出一种表述火焰边界动态形变的方法,首先定位了符合人眼感知的火焰区域中心;其次,利用此中心对火焰边界进行极坐标建模;最后,依据边界模型定义了火焰边界动态形变幅度谱的提取方法,对动态形变过程进行描述,并将其应用于转炉终点的判定。为保证火焰边界动态形变描述的有效性,对原火焰图像及提取到的火焰边界进行了处理。实验结果表明,与现有的差分链码曲率、边界不变矩和圆形度等火焰边界静态特征相比,所提算法有较高的识别率,且能满足实时性要求,有较高应用价值。

转炉炼钢智能控制方法及应用

将模糊推理、自适应、自学习、专家系统等技术与转炉炼钢动态控制方法相结合 ,提出了转炉炼钢智能动态控制的新方法 .该控制方法由动态吹炼过程补吹氧量与冷却剂量预设定模型、熔池碳含量和温度预测模型及停吹专家系统等组成 ,并成功应用于某钢厂转炉炼钢过程 。

基于PSO优化SVM的转炉炼钢用氧量预测研究

用氧量是影响钢水质量的主要因素之一,为提高转炉炼钢用氧量模型的预测精度,提出基于PSO优化SVM的吹氧量建模预测方法。针对SVM结构参数依据经验选取,致使预测模型的泛化能力差,在标准PSO算法的基础上,优化SVM的惩罚系数、不敏感损失系数和高斯核宽度系数3个结构参数,并建立转炉炼钢用氧量预测模型;在此基础上利用UCI数据库中的Auto-MPG标准数据,验证了方法的有效性;最后以某钢厂100 t转炉的实际生产数据建立吹氧量预测模型,结果表明,与标准BP、RBF及SVM相比,基于PSO优化SVM的转炉炼钢吹氧量预测模型精度高、泛化能力强。

基于火焰彩色纹理复杂度特征的转炉炼钢吹炼状态识别

在基于火焰图像识别的转炉吹炼状态识别过程中,针对已有方法存在火焰彩色纹理信息利用不充分和状态识别率仍需提高的问题,提出一种基于火焰彩色纹理复杂度特征的转炉吹炼状态识别方法。首先,将火焰图像转化到HSI颜色空间下并作非均匀量化;然后,计算H分量和S分量的共生矩阵从而融入火焰图像的颜色信息;其次,利用得到的颜色共生矩阵计算火焰纹理复杂度的特征描述子;最后,应用Canberra距离作为相似度度量准则对吹炼状态进行分类和识别。实验结果表明,与已有的转炉火焰灰度共生矩阵和灰度差分统计方法相比,在满足吹炼识别实时性要求的前提下,所提方法的识别率分别提高了28.33%和3.33%。

基于遗传算法和径向基函数神经网络的转炉炼钢模型

针对转炉传统模型的弱点 ,本文在转炉建模过程中引入了遗传算法和径向基函数神经网络 ,由遗传算法辨识转炉过程的脱碳与升温模型 ,并利用径向基函数神经网络及时补偿辨识模型的误差。实际结果表明这一方法效果明显。

基于膜算法进化极限学习机的氧气转炉炼钢终点预报模型

氧气转炉炼钢的控制目标是终点温度和碳含量,但由于不能对其进行在线连续测量,直接影响了出钢的质量.针对该问题,提出一种基于膜算法进化极限学习机(ELM)的抗干扰终点预报模型.利用进化膜算法的全局寻优能力调整ELM网络参数,不仅避免了ELM网络受异常点影响出现过拟合现象,还可以寻找最优复杂度的ELM模型.将找到的ELM模型应用到转炉炼钢领域并建立终点碳含量和温度的预报模型.在仿真实验中,分别使用含有高斯噪声的标准sin C函数和氧气转炉炼钢实际生产数据进行仿真,结果表明所提模型在含噪声的数据中具有较好的预报精度和鲁棒性.

基于支持向量机碱度偏差估计的石灰加入量模型

提出碱度偏差估计模型,进而提出一种转炉炼钢石灰加入量计算方法.首先,使用支持向量机建立碱度偏差估计模型,预报输入碱度与炉渣碱度之间的偏差;然后,利用碱度偏差估计模型预报的偏差值修正传统公式中的碱度参数,再消除白云石加入量对石灰加入量的影响,得到改进的石灰加入量模型.应用该方法对一座150t转炉的实际生产数据进行计算,结果显示对碱度偏差的预报有着较高的精度,由此计算的石灰加入量可以满足实际生产的要求.

氧气转炉煤气全干法显热回收系统中CO爆燃与防爆研究

氧气转炉煤气一般在850°C左右时采用喷水/水雾法降温除尘,导致煤气50%的显热被浪费.为了充分利用转炉炼钢过程中富含CO煤气的余热资源,新方法取消了喷水工艺,采用转炉煤气全干法显热回收系统,但是该技术在转炉煤气前烧与后烧阶段存在煤气爆炸的风险.针对转炉全干法系统的安全稳定运行需求,通过实验和理论计算研究了CO当量比、混合气初始温度和含水量等因素对CO爆燃特性的影响.结果表明:CO爆燃的最大压力和火焰速度随着混合气体中CO当量比的减小呈现减少的趋势,但当CO当量比小于0.368时,则对火焰速度的影响不大.在实验CO当量比范围内,爆燃压力最大值为0.65 MPa,最大爆燃速度约为750 m/s;混合气体初始温度升高导致爆燃过程中产生的最大爆燃压力降低,与此同时火焰速度会相对增加,进而影响火焰传播时间.含水量增加会导致CO爆燃的最大爆燃压力的增加,但含水量到达0.463%后继续增大则对最大爆燃压力影响不大;最后,通过分析CO爆燃特性和实际生产过程,提出了燃烧控制与强化以及煤气爆炸遏制等防爆方法和技术,从而有效降低爆燃带来的损失.

基于氧气脱碳效率预测的转炉炼钢吹氧量计算模型

为精确计算转炉炼钢生产过程中需要吹入的氧气量,提出了基于氧气脱碳效率预测的转炉炼钢静态和动态吹氧量计算模型.首先,采用独立成分分析方法对静态模型输入进行预处理;然后,建立基于支持向量机的氧气脱碳效率预测模型;最后,利用预测得到的氧气脱碳效率结合机理公式计算两阶段吹氧量.利用一座150t转炉的实际生产数据进行仿真计算,结果显示该模型对氧气脱碳效率的预报精度较高,所提方法是有效的.

基于粗糙集的转炉炼钢知识发现模型

针对转炉炼钢知识发现的特点,采用粗糙集理论进行分析,应用数据清洗、标准化及离散等方式对转炉炼钢生产数据进行预处理,以炼钢生产的主要影响因素作为知识发现的条件属性,以转炉冶炼终点控制目标作为知识发现的决策属性,建立了基于粗糙集方法的转炉炼钢知识发现模型,实现转炉炼钢生产知识的自动发现、获取和规则提取。以转炉冶炼终点钢水温度的变化规律做为知识发现的决策属性,采用210t转炉炼钢实际生产数据进行模型的应用测试,结果表明提取出的铁水硅含量、铁矿石质量、氧气消耗量等影响因素对转炉冶炼钢水终点温度存在重要影响,且模型提取出的转炉炼钢终点钢水温度知识规则与现行转炉炼钢现场的变化规律一致,证明基于粗糙集方法的转炉炼钢知识发现模型的有效性。

论宝钢转炉的经济炉龄

研究了宝钢转炉的综合经济炉龄。