高炉专家系统在太钢高炉的应用

对太钢高炉专家系统作了简单介绍,分析了太钢高炉专家系统的主要功能,重点阐述了应用效果较好的配料模型、热风炉优化/控制模型、炉身仿真模型和炉缸侵蚀模型的应用情况,根据高炉专家系统运行期间发生的问题,提出了影响高炉专家系统运行的主要因素,并结合应用提出了开发高炉专家系统的一些建议。     

本钢炼铁厂5号高炉专家系统应用实践

针对本钢开发高炉专家系统的背景和高炉专家系统的概念作了简单介绍,分析了本钢5号高炉专家系统的结构、软硬件配置情况及其应用功能,阐述了配料模型、渣铁平衡模型和炉缸侵蚀模型的应用,根据高炉专家系统运行期间发生的问题,提出了影响高炉专家系统运行的主要因素,并结合应用提出了开发高炉专家系统的一些建议。     

太钢炼铁厂五高炉专家系统应用实践

对太钢开发高炉专家系统的背景和高炉专家系统的概念作了简单介绍,分析了太钢五高炉专家系统的结构、软硬件配置情况及其应用功能,阐述了配料模型、渣铁平衡模型和炉缸侵蚀模型的应用,根据高炉专家系统运行期间发生的问题,提出了影响高炉专家系统运行的主要因素,并结合应用提出了开发高炉专家系统的一些建议。     

马钢新1号高炉平衡点操作技术的研究

对马钢新１＃高炉自９５年以来的数据进行了分析，采用人工神经网络方法设立了一和中高炉下部热量与高炉各主要参数间的预报神经网络模型。该模型可进行瞬时预报高炉下部热量水平，并找出与平衡点之间的偏离度，从而调整各参数，使高炉逼近平衡点，达到稳定高炉操作目的，该模型还具有良好的适应性和自学习功能。     

铜冷却壁高炉操作炉型诊断管理模型的开发与应用

利用自组织特征映射人工神经网络理论(SOFM)，开发了一个高炉操作炉型诊断管理模型。通过将该模型用于分析武钢1号高炉的炉型管理，发现利用该模型可以及时分析高炉操作炉型的波动及变化，从而帮助高炉操作人员调整炉况，维护高炉的稳定顺行。     

冶金企业中高炉系数目标优化模型构建分析

基于冶金行业的快速发展,高炉模型的优化成为重点问题。在生产过程中,高炉参数代表着其生产率的高低。所以,关于冶金企业中高炉系数目标优化模型构建分析成为了目前研究主题。以高炉利用系数为目标,构建高炉物理模型,然后进行优化变量的选择工作,以及高炉系数求值,最后制定了包括工艺、操作、变量等50个约束条件,并以此为基础构建整个优化模型,利用序列二次规划算法分析最优结果。通过实验进行论证,证明高炉系数目标优化模型的构建在应用中具有良好的效果。     

高炉无钟布料模型的改进与发展

探讨了目前国内高炉无钟布料模型的现状,介绍了首钢高炉布料模型中的改进和创新环节,提出了今后布料模型的发展方向,并通过首钢二高炉规范化布料调整试验分析了其可行性.     

基于支持向量机的铁水硅含量的预测

为了准确预测高炉炼铁过程的硅含量,分析了高炉工艺参数对高炉铁水硅含量的时序性影响,以支持向量机理论为基础构建了2类铁水硅含量预测模型,即硅含量模型和硅变化量模型。利用首钢迁钢3号高炉铁水硅含量数据进行模型测试,测试结果表明2类模型预测命中率均可达到80%。     

KTH高炉模拟和预报数模的开发

开发了自料线至风口的一维静态高炉数模——KTH高炉模拟和预报模型。模型描述了炉内的变化行为,模拟了实际的高炉过程,有助于加深对高炉理论的理解,也可在提高风温、富氧、喷煤、富氧喷煤、改善焦炭质量等情形下预报高炉操作指标和炉况的变化。     

高炉理论燃烧温度计算模型的开发与实践

在传统理论燃烧温度计算模型的基础上,重点考虑了风口回旋区燃烧平衡时煤气成分的变化,强调化学平衡对理论燃烧温度的影响,提出了新的理论燃烧温度计算模型,通过采集与处理现场高炉数据以及编程技术实现了该模型的在线计算,并成功应用于国丰钢铁有限公司1号高炉,在高炉与主控室之间建立起一条理论燃烧温度信息通道,实时反映炉缸热状态。实践证明,该模型可以有效帮助工长了解高炉情况,从而对于保持炉缸状态活跃、高炉顺行、延长高炉寿命等起到了促进作用。     

LF炉精炼温度预报模型的发展应用

全面综述了LF炉精炼温度预报模型在国内外的发展应用状况。国内外所开发的LF炉钢水温度预报模型,主要采用机理模型、“黑箱模型”、“灰箱模型”三种方法。简单讨论了这三种模型的特性,并预测了LF炉精炼温度模型的发展趋势。     

典型离子化合物标准熵的估算模型

 提出了用于估算典型离子化合物标准熵的双参数模型。此模型可以用来估算由碱金属、碱土金属与氧族元素和卤族元素组成的离子化合物的标准熵。模型通过回归求解给出了估算时各元素的估算参数值。由此模型估算离子化合物的标准熵时,平均误差为2.473J·mol-1·K-1,标准差为3.274 J·mol-1·K-1。模型估算误差符合正态分布,并且在与离子束缚模型等估算模型比较中显示了较高的精确度,优于现有的其他估算方法。     

Hybrid Modeling for Soft Sensing of Molten Steel Temperature in LF

 Aiming at the limitations of traditional thermal model and intelligent model, a new hybrid model is established for soft sensing of the molten steel temperature in LF. Firstly, a thermal model based on energy conservation is described; and then, an improved intelligent model based on process data is presented by ensemble ELM (extreme learning machine) for predicting the molten steel temperature in LF. Secondly, the self adaptive data fusion is proposed as a hybrid modeling method to combine the thermal model with the intelligent model. The new hybrid model could complement mutual advantage of two models by combination. It can overcome the shortcoming of parameters obtained on line hardly in a thermal model and the disadvantage of lacking the analysis of ladle furnace metallurgical process in an intelligent model. The new hybrid model is applied to a 300 t LF in Baoshan Iron and Steel Co Ltd for predicting the molten steel temperature. The experiments demonstrate that the hybrid model has good generalization performance and high accuracy.     

Rasch analysis of distractors in multiple-choice items

In order to apply the Rasch model to multiple-choice items, incorrect responses to distractors are usually aggregated to a single category. In doing so, information of individual distractors disappears. In this paper, a Rasch-type analysis is proposed where one parameter is assigned to each distractor. The information is thus preserved. The proposed distractor model can be applied to investigate the performance of distractors, which is useful for item revision. This model is a necessary condition of the Rasch model, that is, fitting the distractor model will fit the Rasch model, but not vice versa. The results of a small simulation study show that parameter recovery of the distractor model is very satisfactory. A real data set of twenty multiple-choice items was analyzed. Some items were found to fit the Rasch model rather than the distractor model. It is this diagnostic value that makes the distractor model suitable for multiple-choice items.     

Identification and Modeling of Permanent Deformations of Asphalt Concrete

The accumulation of permanent deformations in asphalt concrete under repeated loading, a phenomenon of great practical importance, is studied in detail and a new multidimensional nonlinear hyperelastic-nonassociative plastic model is proposed. The equations of the model are solved by using the predictor–corrector algorithm. The model’s results are compared to actual experimental data and the comparison is very favorable. A remarkable result is that the model parameters obtained from one stress level predict correctly the results for another stress level.     

Learning Curve Models of Construction Productivity

Current research into learning curve models of construction productivity is presented. Five mathematical models are identified and each of these are used to model unit rates for 65 sets of data. The correlation between predicted and actual unit rates is determined, and on this basis, it is concluded that the best predictor is a cubic model. The often cited straight‐line model is only marginally adequate. The validity of the straight‐line model is further undermined by showing that the learning rate is not a constant value. Time data for erecting and setting 466 precast concrete floor planks is used to support the conclusion that the straight‐line model is not a reliable model for predicting future performance.     

Simulation of Bed-Load Dispersion Process

Two general dispersion models suitable for nonequilibrium bed-load transport were constructed. The first one, called the P model, is based on the probability of migration for specific groups of sediment particles. The second one, called the D model, is derived from the advection equation discretized in finite-difference form, which is equivalent to the general dispersion equation. By comparing these models, it is found that the D model can be treated like the P model in some respects. The Courant number, Cr, in the D model has the same physical meaning as the probability of migration, P, in the P model. Although the D model and P model were based on different concepts, the simulated bed-load transport rates, which result from their application, are the same. Therefore, the dispersion equation was replaced by the numerical algorithm of the advection equation (D model) to examine several dispersion phenomena of bed-load transport. To explore further the nonequilibrium dispersion process, a series of flume experiments was conducted by using color-painted fine gravels. Having compared model simulation results and experimental data, it is shown that the models derived in this study have a reasonably good agreement with the experimental results. In summary, this study has indirectly proven that the D model, which is equivalent to the dispersion equation, is capable of simulating the nonequilibrium bed-load transport.     

铸钢件并行开发模式的研究

铸钢件是市场前景非常好的钢制半成品。它在传统的开发模式下，开发周期长，开发成本高，开发人员彼此间协同差，非常不适合现在快速响应顾客需求的市场经济。基于此，提出了一种新的开发模式——铸钢件并行开发模式，并对该模式的内容、关键技术进行了讨论。     

结构加速度频响函数模型修正的Kriging方法

为提高结构频响函数模型修正效率,提出将Kriging模型引入优化过程,代替有限元模型进行迭代运算.基于频响曲线对应频率点处的响应值之差构造目标函数,并结合初选设计参数进行实验设计.根据实验设计结果进行各参数的灵敏度分析,进而筛选出模型修正的待修正参数,基于该参数及其响应构造Kriging模型,经检验有效的Kriging模型将参与模型修正过程.以GARTEUR飞机模型为算例,基于加速度频响数据进行模型修正,修正后模型不仅能复现检验点处频响曲线,还能成功预测结构局部修改后的频响曲线,证明了Kriging方法应用于频响函数模型修正的有效性.      

变尺度混沌神经网络在球磨机故障诊断中的应用

针对神经网络极易陷入局部极小的问题,提出变尺度混沌优化方法与神经网络相结合的变尺度混沌神经网络综合模型,讨论变尺度混沌神经网络模型的建模思路及其算法实现;介绍球磨机常见的故障类型,建立球磨机故障诊断的变尺度混沌神经网络模型,并进行仿真试验,结果表明:该模型具有较高的预测精度,可以有效地运用于球磨机诊断中的故障预测。