Endpoint temperature prediction of molten steel in RH using improved case-based reasoning

An improved case-based reasoning (CBR) method was proposed to predict the endpoint temperature of molten steel in Ruhrstahl Heraeus (RH) process. Firstly, production data were analyzed by multiple linear regressions and a pairwise comparison matrix in analytic hierarchy process (AHP) was determined by this linear regression’s coefficient. The weights of various influencing factors were obtained by AHP. Secondly, the dividable principles of case base including “0–1” and “breakpoint” were proposed, and the case base was divided into several homogeneous parts. Finally, the improved CBR was compared with ordinary CBR, which is based on the even weight and the single base. The results show that the improved CBR has a higher hit rate for predicting the endpoint temperature of molten steel in RH.     

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 prediction 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.     

Predicting the martensite transformation start-temperature of low alloy steel based on fuzzy identification

A method of fuzzy identification based on T-S fuzzy model was proposed for predicting temperature M_s from chemical composition, austenitizing temperature and time for low alloy steel. The degree of membership of each sample was calculated with fuzzy clustering algorithm. Kalman filtering was used to identify the consequent parameters. Compared with the results obtained by empirical models based on the same data, the results by the fuzzy method showed good precision. The accuracy of the fuzzy model is almost 6 times higher than that of the best empirical model. The influence of alloying elements, austenitizing temperature and time on M_s was analyzed quantitatively by using the fuzzy model. It is shown that there exists a nonlinear relationship between the contents of alloying elements in steels and their M_s, and the effects of austenltizing temperature and time on M_s temperature cannot be neglected.     

Effect of initial perturbations on the simulation capability of numerical models

The effect of an initial perturbation on simulation is studied in a series of numerical experiments using a one-dimensional nonlinear advection equation model. It is shown that the capability of the model tends to decline with an increase in initial perturbation. The error in the initial perturbation grows in a nonlinear manner, indicating that the simulating results can be improved only to a limited degree by increasing the accuracy of the initial data.     

Effects of calculation approaches for thermal conductivity on the simulation accuracy of billet continuous casting

An unsteady, two-dimensional, explicitly solved finite difference heat transfer model of a billet caster was presented to clarify the influence of the thermal conductivity of steel on model accuracy. Different approaches were utilized for calculating the thermal conductivity of solid, mushy and liquid steels. Model results predicted by these approaches were compared, and the advantages of advocated approaches were discussed. It is found that the approach for calculating the thermal conductivity of solid steel notably influences model predictions. Convection effects of liquid steel should be considered properly while calculating the thermal conductivity of mushy steel. Different values of the effective thermal conductivity of liquid steel adopted could partly be explained by the fact that different models adopted dissimilar approaches for calculating the thermal conductivity of solid and mushy steels.     

Dynamic prediction of building subsidence deformation with data-based mechanistic self-memory model

This paper describes a building subsidence deformation prediction model with the self-memorization principle.According to the non-linear specificity and monotonic growth characteristics of the time series of building subsidence deformation,a data-based mechanistic self-memory model considering randomness and dynamic features of building subsidence deformation is established based on the dynamic data retrieved method and the self-memorization equation.This model first deduces the differential equation of the building subsidence deformation system using the dynamic retrieved method,which treats the monitored time series data as particular solutions of the nonlinear dynamic system.Then,the differential equation is evolved into a difference-integral equation by the self-memory function to establish the self-memory model of dynamic system for predicting nonlinear building subsidence deformation.As the memory coefficients of the proposed model are calculated with historical data,which contain useful information for the prediction and overcome the shortcomings of the average prediction,the model can predict extreme values of a system and provide higher fitting precision and prediction accuracy than deterministic or random statistical prediction methods.The model was applied to subsidence deformation prediction of a building in Xi’an.It was shown that the model is valid and feasible in predicting building subsidence deformation with good accuracy.     

Effect of magnesium and nickel coatings on the wetting behavior of alumina toughened zirconia by molten Al-Mg alloy

The wettability of alumina toughened zirconia (ZTA) by Al-Mg alloy was investigated using the sessile drop technique. The effects of nickel coating, magnesium content, nitrogen atmosphere, and processing temperature on the contact angle between the molten alloy and the substrate were determined. Likewise, the effect of these factors on the wetting properties was studied. The results showed that the nickel coating on the ceramic substrate caused a significant reduction in solid/liquid surface energy and the contact angle decreased obviously. The presence of magnesium in the molten aluminum alloy in nitrogen atmosphere reduced the contact angle effectively. The presence of magnesium in the alloy must be at a minimum amount of 2wt%-3wt%. Moreover, it was suggested that some chemical reactions in the Al-Mg-N system led to the production of Mg₃N₂ and AlN compositions. These compositions improved the wetting properties of the systems by reducing the surface energy of the molten. It was shown that increasing the temperature is also an effective factor for the enhancement of wetting properties.     

Precipitation behavior of BN type inclusions in 42CrMo steel

Automobile crankshaft steel 42CrMo, which requires excellent machinability and mechanical properties, cannot be manufactured by traditional methods. To achieve these qualities, the formation behavior of boron nitride (BN) inclusions in 42CrMo steel was studied in this article. First, the precipitation temperature and the amount of BN type inclusions with different contents of boron and nitrogen in molten steel were calculated thermodynamically by FactSage software. Then the morphology and the size of BN type inclusions as well as the influence of cooling methods on them were investigated by scanning electron microscopy. Furthermore, the effects of cooling rate and the contents of B and N in molten steel on the morphology, size, and distribution of BN type inclusions were studied quantitatively and detailedly by directional solidification experiments. It is found that different BN inclusions in molten steel can form by controlling the cooling rate and the contents of B and N, which is important for obtaining the excellent machinability of 42CrMo steel.     

Fashion Color Forecasting by Applying an Improved Back Propagation Neural Network

Fashion color forecasting is one of the most important factors for fashion marketing and manufacturing. Several models have been applied by previous researchers to conduct fashion color forecasting. However, few convincing forecasting systems have been established. A prediction model for fashion color forecasting was established by applying an improved back propagation neural network (BPNN) model in this paper. Successive six-year fashion color palettes, released by INTERCOLOR, were used as learning information for the neural network to develop a reliable prediction model. Colors in the palettes were quantified by PANTONE color system. Additionally, performance of the established model was compared with other GM(1, 1) models. Results show that the improved BPNN model is suitable to predict future fashion color trend.     

Satellite orbit determination by transfer with differenced ranges

The original idea of orbit determination called "determination of satellite orbit by transfer" was proposed by the National Time Service Center,Chinese Academy of Sciences.It shows that the system is very stable and the orbit determination accuracy is improved greatly.A new observation mode called "differenced ranges between master station and slave stations by transfer" is introduced.It is the development of "determination of satellite orbit by transfer".In principle,the differenced ranges between master station and slave stations have a high angular resolution,and strongly constrains in the transverse direction of satellite orbit,perpendicular to the line-of-sight.The principle of "differenced ranges between master station and slave stations by transfer" is discussed in detail.And differenced ranges in combination with ranging data were used to determine satellite’s orbit and orbit prediction under different arc observations.It shows that orbit determination accuracy for orbit prediction can be improved with differenced ranges in combination with ranging data.     

Development of an improved CBR model for predicting steel temperature in ladle furnace refining

In the prediction of the end-point molten steel temperature of the ladle furnace, the influence of some factors is nonlinear. The prediction accuracy will be affected by directly inputting these nonlinear factors into the data-driven model. To solve this problem, an improved case-based reasoning model based on heat transfer calculation(CBR-HTC) was established through the nonlinear processing of these factors with software Ansys. The results showed that the CBR-HTC model improves the prediction accuracy of end-point molten steel temperature by5.33% and 7.00% compared with the original CBR model and 6.66% and 5.33% compared with the back propagation neural network(BPNN)model in the ranges of [-3, 3] and [-7, 7], respectively. It was found that the mean absolute error(MAE) and root-mean-square error(RMSE)values of the CBR-HTC model are also lower. It was verified that the prediction accuracy of the data-driven model can be improved by combining the mechanism model with the data-driven model.     

基于智能技术的温度测量系统的设计

钢水温度是炼钢过程的重要控制指标。目前由于钢水温度过高和钢液、钢渣对测温枪的腐蚀,现有的测量温度的方法无法得到钢水温度的连续变化的信息。针对这种情况,介绍了基于智能技术的钢水温度的软测量方法。应用人工神经元网络BP算法对钢水温度进行初步预报,再根据专家系统知识对一些特殊的情况进行修正,从而获得良好的效果。运行结果表明:采用该测量方法,钢水温度预报的适应性和准确性都得到了显著提高,并且该方法在安全性、操作控制及经济效益等方面具有很大的优越性,具有广阔的发展前景。     

基于灰箱模型的LF精炼终点温度预测

以国内某钢厂一30 t LF精炼炉为研究对象,通过建立由传热机理模型和黑箱模型相结合的灰箱模型对LF精炼终点温度进行了预测。首先根据能量守恒定律建立了传热机理模型。针对包衬耐材的蓄热以及合金的热效应难以精确计算的问题,采用偏最小二乘黑箱模型对这一部分温度进行了处理,最后将两种模型相结合综合预测了LF钢包精炼的终点温度。结果表明,偏最小二乘法在预测包衬的耐材蓄热和合金热方面的温度误差在±5℃以内的命中率达到97%以上,总的灰箱模型预测LF精炼终点温度误差在±5、±8、±10℃以内的命中率分别达到88%、96%和99%,模型具有较高的预测精度。研究可为该钢厂的LF精炼工艺提供指导。     

考虑光伏机理与数据驱动结合的短期功率预测

精确的光伏功率预测对电网的可靠与稳定运行至关重要。现有研究大多数都是将天气条件直接作为数据驱动的输入,未深入分析天气条件与光伏输出功率直接耦合关系,因此本文将机理模型与数据驱动方法相结合,提出一种新型的光伏功率预测方法。首先,建立光伏系统物理模型,依据建立的物理模型得到不同的辐照度分量以及光伏电池温度。其次,将这些关键的天气特征重新构建数据驱动的输入,实现光伏机理与数据驱动结合的短期功率预测。最后,进行误差修正然后得到最终的光伏功率预测结果。根据光伏系统实测数据集进行仿真分析,结果表明因为从物理模型得到了关键天气特征,考虑了天气条件与天气因素的耦合关系,预测精度有了明显提升,验证了所提方法的有效性。     

基于主元分析的转炉终点ESN静态预测模型

以主元分析方法和新型ESN(回声状态网络)算法为核心,研究了转炉终点静态预测模型。通过对某钢厂转炉生产数据的主元分析,建立了ESN模型,同时将ESN模型与传统的BP和RBF神经网络模型进行了对比研究。结果表明,使用ESN建立的模型比传统的BP网络模型和RBF网络模型,在钢水温度预测方面精度分别提高了0.85%和0.45%,在钢水碳质量分数预测方面精度分别提高了0.45%和0.19%,能够有效的对转炉终点碳含量和温度进行预测,从而为转炉炼钢过程提供更准确的操作指导。     

基于改进灰色模型的MEMS陀螺仪零偏温度补偿

微机电系统(micro-electro mechanical system,MEMS)陀螺仪的零点漂移是影响陀螺仪测量精度的主要因素.针对MEMS陀螺仪零点漂移随温度变化的非线性问题,以MEMS惯性传感器为试验对象,采用小波变换对MEMS陀螺静态实验零偏数据进行滤波,结合改进灰色预测模型估计零偏随温度变化趋势,获得基于小波变换和改进灰色预测的温度补偿模型.与常规补偿模型算法比较表明,基于小波变换和改进灰色预测的温度补偿模型均方根误差和平均绝对误差更小,MEMS陀螺仪零点漂移的均方根误差和平均绝对误差分别减少到0.025 0和0.018 0,验证了该补偿模型的可行性,对提高陀螺测量精度具有较好的理论意义和工程应用价值.     

广义相加模型在乌江夏季径流预报中的应用

基于前期冬季海温指数,构建具有4个非线性指数和9个线性指数的广义相加模型(GAM),对乌江流域洪家渡夏季径流进行了模拟与预测,利用5种评估指标,包括最小信息准则(AIC)、均方根误差(RMSE)、平均绝对误差(MAE)、概率空间线性误差(LEPS)和线性相关(r),评估GAM和广义线性模型(GLM)模拟效果.结果表明:...     

计及机理机制的Stacking集成光伏发电预测

准确地光伏预测对电力调度、容量分析和机组组合至关重要。现有的数据驱动预测算法在计算速度和预测精度上有一定的提升,但未能考虑光伏发电的内在机理,存在泛化的风险。针对上述问题,提出了一种基于Stacking框架的机理模型和数据驱动结合的预测模型。其中,光伏发电机理模型将嵌入Stacking框架一层预测结构,构成基于长短期记忆神经网络(long short-term memory, LSTM)、极度梯度提升树(extreme gradient boosting, XGBoost)和机理模型的并行预测学习器。机理模型将光伏发电限制在一个合理的范围内,作为数据驱动模型的预测约束。所提出的模型能够从机理模型中提取有用的固有信息,并利用数据分析的能力提取历史数据中的非线性关系。基于安徽省某地区实际数据分析,所提模型相比传统数据驱动方法具有更高的精度。     

基于粒子群-投影寻踪和遗传-神经网络集成的预测模型

  针对预测对象和预测因子存在复杂的线性和非线性关系的特点,利用自然正交展开方法进行线性降维,以及用粒子群 投影寻踪方法进行非线性降维,将高维的非线性数据投影到低维子空间上,构造了一种遗传 神经网络预测模型。在此基础上,应用该预测模型对影响华南的台风频数进行了预测试验,并将预测结果与统计回归模型的预测结果进行对比分析。结果表明,文中构建的非线性集预测模型,对台风频数有较好的预测效果,5 年预测的平均绝对误差为0.81个, 平均相对误差为13%,预测结果比统计回归模型有明显的改进。该文的结果可为进一步探索研究其他领域的预测建模提供了一种新的参考思路和方法。