Experimental measurements of load and strip profile in thin strip rolling

Experiments are described in which plasticine strips are rolled using elastomer rolls. Conditions cover the range from “thick strip” behaviour, in which roll elastic deformations are small, to “thin strip rolling”, in which elastic deformations of the rolls are very significant. Results provide the first direct experimental confirmation of the thin strip rolling model proposed by Fleck et al. (Proc. Instn Mech. Engrs B206 (1992) 119–131). Strip profiles clearly show a short region of reduction at the inlet to the bite and a central region which is relatively flat, in accord with the theory. The profiles do not however show a short region of reduction at the exit as predicted. For intermediate strip thicknesses the measured loads are in reasonable agreement with theory. For the thinnest strips, although the form of the dependence of load on reduction and inlet strip thickness is as predicted by theory, the measured loads are almost an order of magnitude lower than predicted. It is suggested that this is caused either by differences between the assumed rigid–perfectly plastic strip and the real constitutive behaviour of the plasticine, or by errors in treating the rolls as elastic half-spaces, an approximation which is accurate for industrial metal rolling, but is not good for the conditions of these experiments.     

成形法制备半蜂窝波形条的有限元模拟

航空工业上采用的以成形法成形半蜂窝波形条的方式,成形的波形条结构均匀、精度高、工艺间单.根据某厂生产波形条的工艺,利用非线性有限元法开发了成形法的有限元模型,运用该模型对现场的实际成形过程进行仿真计算,得到与现场实测相吻合的结果.比较板坯在单、双主动辊加栽的条件下,成形结果的优劣性.在此基础上,对影响成形结果的主要因素进行分析.     

Study on the improved accuracy of strip profile using numerical formula model in continuous cold rolling with 6-high mill

The quality requirements for thickness accuracy in cold rolling continue to become more stringent. In cold rolling mill, it is very important that the rolling force calculation considers rolling conditions. The rolled strip thickness was predicted using calculated rolling force. However, the prediction of strip thickness in cold rolling is very difficult; in particular, for 6-high mill with shifted intermediate roll (IMR), the accuracy of thickness is not good. In this study, to improve the accuracy of rolled strip thickness, the roll gap flattening can be given based on Hertz contact theory, with contact between rolls and the smooth cylindrical rolls for the rolling elastic deformation. Also, the distribution of the roll gap flattening may be calculated using the contact force of unit transverse length. The strip profile at the continuous cold rolling is calculated by using the numerical analysis model considering the initial strip profile before cold rolling. Hence, we propose that the numerical model can predict the rolled strip profile more quickly and accurately and be applicable to the field. The results of the proposed numerical model were verified by FE-simulation and cold rolling experiments of 6-high mill with five stands.     

基于点云数据重构三维网格模型的神经网络法

给出了一种新的基于神经网络的点云数据重构CAD实体模型的新方法。该方法能直接从神经网络的权值矩阵得到曲线的控制顶点、曲面的控制网格，通过神经网络的权值约束实现曲线段、曲面片之间的光滑拼接。同时对恢复的隐式表面的初始逼近网格自适应性进行优化。利用谊方法恢复的网格形状，无论在表面还是在隐式曲面的轮廓部分都能获得很好的视觉效果。所有算法的时间复杂度均为O（n），可以完全实时进行。     

几种典型板带轧机板型控制技术简介

带钢板型是冷轧板带及其涂镀产品最重要的质量指标,近年来出现了一大批可选择的具有良好板型控制功能的轧机和技术,本文将对UCM、CVC、Smart—Crown、DSR及六辊3C这几种新型轧机的板型控制技术做一下简要介绍。     

板带轧制有限元建模研究

基于大型通用有限元软件MSC.Marc,提出了对板带轧制模型进行网格划分的方法,并以板带轧机辊型设计为研究背景,具体介绍了其网格划分的步骤和方法.通过多个网格划分模型进行对比,证明文中提出的网格划分方法具有准确性和广泛性的特点,为同行业的有限元模拟人员提供了一种分析建模的思路和方法.     

Road profile estimation using neural network algorithm

This paper more specifically focuses on the estimation of a road profile (i.e., along the “wheel track”). Road profile measurements have been performed to evaluate the ride quality of a newly constructed pavement, to monitor the condition of road networks in road management systems, as an input to vehicle dynamic systems, etc. The measurement may be conducted by a slow-moving apparatus directly measuring the elevation of the road or using a means that measures surface roughness at highway speeds by means of accelerometers coupled with high speed distance sensors, such as laser sensors or using a vehicle equipped with a response-type road roughness measuring system that indirectly indicate the user’s feelings of the ride quality. This paper proposes a solution to the road profile estimation using an artificial neural network (ANN) approach. The method incorporates an ANN which is trained using the data obtained from a validated vehicle model in the ADAMS software to approximate road profiles via the accelerations picked up from the vehicle. The study investigates the estimation capability of neural networks through comparison between some estimated and real road profiles in the form of actual road roughness and power spectral density.     

Neural network prediction of disc brake performance

An automotive brake's performance results from the complex interrelated phenomena occurring at the contact of the friction pair. These complex braking phenomena are mostly affected by the tribochemical properties of the friction material's ingredients, the brake disc properties, and the brake's operating regimes. In this paper, the synergistic effects of the friction material's properties, defined by its composition and manufacturing conditions, and the brake's operating regimes on the disc brake factor C variation have been modelled by means of artificial neural networks. The influences of 26 input parameters, determined by the friction material composition (18 ingredients), its manufacturing conditions (5 parameters), and the brake's operating regimes (3 parameters) on the brake factor C variation, have been predicted. The neural model of the disc brake cold performance has been developed by training 18 different neural network architectures with the five different learning algorithms. The optimal neural model of disc brake operation has been shown to be valid for predicting the brake factor C variation of the cold disc brake over a wide range of brake's operating regimes and for different types of friction material.     

Adaptive neural network controller for the molten steel level control of strip casting processes

The twin-roll strip casting process is a steel-strip production method which combines continuous casting and hot rolling processes. The production line from molten liquid steel to the final steel-strip is shortened and the production cost is reduced significantly as compared to conventional continuous casting. The quality of strip casting process depends on many process parameters, such as molten steel level in the pool, solidification position, and roll gap. Their relationships are complex and the strip casting process has the properties of nonlinear uncertainty and time-varying characteristics. It is difficult to establish an accurate process model for designing a model-based controller to monitor the strip quality. In this paper, a model-free adaptive neural network controller is developed to overcome this problem. The proposed control strategy is based on a neural network structure combined with a sliding-mode control scheme. An adaptive rule is employed to on-line adjust the weights of radial basis functions by using the reaching condition of a specified sliding surface. This surface has the on-line learning ability to respond to the system’s nonlinear and time-varying behaviors. Since this model-free controller has a simple control structure and small number of control parameters, it is easy to implement. Simulation results, based on a semiexperimental system dynamic model and parameters, are executed to show the control performance of the proposed intelligent controller. In addition, the control performance is compared with that of a traditional PID controller.     

神经网络技术在称重传感器中的应用

通过实际模型及其分析提出了几种快速称重方法和提高传感器性能的原理,着重阐述了基于神经网络理论,设计出称重传感器神经网络动态补偿器和提高传感器性能。仿真研究表明有效的提高了传感器动态响应的快速性,且对系统参数变化具有鲁棒性。将先进控制理论应用于实际传感器系统中,为提高传感器性能开辟了新途径,促进了称重传感器行业的发展。     

Neural network controller for robotic motion control

Since a robotic manipulator has a complicated mathematical model, it is difficult to design a control system based on the complicated multi-variable nonlinear coupling dynamic model. Intelligent controllers using fuzzy and neural network approaches do not need a real mathematical model to design the control structure and have attracted the attention of robotic control researchers recently. A traditional fuzzy logic controller does not have learning capability and it needs a lot of effort to search for the optimal control rules and the shapes of membership functions. Owing to the time-varying behaviour of the system, the required fine tracking accuracy is difficult to achieve by adjusting the fuzzy rules only. The implementation problems of neural network control are the initial training and initial transient stability. In order to improve the position control accuracy and system robustness for industrial applications, a neural controller is first trained off-line by using the input and output (I/O) data of a traditional fuzzy controller. Then the neural controller is implemented on a five-degrees-of-freedom robot with a back propagation algorithm for online adjustment. The experimental results show that this neural network controller achieved the required trajectory tracking accuracy after 15 on-line operations.     

Neural network approach to ultrasonic flow measurements

Mathematical models and numerical methods offer a flexible tool to investigate flow disturbance effects on flowmeters of different types. In this paper a simple neural network based approach has been used to study the velocity profile dependence of ultrasonic flowmeters. Neural networks have been used in two ways: to interpolate the velocity profiles in the points needed for the modelling of ultrasonic flow measurement, and to compute the weights for different paths of multipath ultrasonic flowmeters. In the former case two types of neural networks, multilayer perceptron networks and radial basis function networks, have been investigated. In the latter case, a single layer neural network with linear neurons is first trained with known velocity profiles, and the weights determined by the network have then been used in the computation of the errors in other piping configurations. The results have been compared with the errors computed with the weights for different paths given in Pannel CN, Evans WAB, Jackson DA. A new integration technique for flowmeters with chordal paths, Flow Measurement and Instrumentation 1990;1:216–224.     

Neural network prediction of brake friction materials wear

Wear of brake friction materials depends on many factors such as temperature, applied load, sliding velocity, properties of mating materials, and durability of the transfer layer. Prediction of friction materials wear versus their formulation and manufacturing conditions in synergy with brakes operating conditions can be considered as a crucial issue for further friction materials development. In this paper, the artificial neural network abilities have been used for predicting wear of the friction materials versus influence of all relevant factors. The neural model of friction materials wear has been developed taking into account: (i) complete formulation of the friction material (18 ingredients), (ii) the most important manufacturing conditions of the friction material (5 parameters), (iii) applied load and sliding velocity of the friction material both represented by work done by brake application, and (iv) brake interface temperature.     

A robust model for rolling of thin strip and foil

A new analysis for cold rolling of thin strip and foil is developed. This model follows the approach of Fleck et al. [8], but relaxes their assumption of a central flat neutral zone. Instead of following their inverse method to obtain the pressure distribution in this neutral zone, an explicit equation for the contact pressure variation is obtained from the sticking condition in this region. This significantly simplifies the solution method, leading to a much more robust algorithm. Moreover the method treats the cases either where the roll retains its circular arc or where there is very significant roll deformation in the same way, greatly simplifying the method of obtaining solutions. This will facilitate the incorporation of other effects such as the friction models currently being developed. Results are in line with the theory of Fleck et al. [8]. The effect of entry and exit tensions on the non-dimensional load and forward slip is investigated. It is found that the effect of equal entry and exit tensions is equivalent to reducing the yield stress of the strip by this tension stress.     

Optimization of the cooling regime of hot-rolled complex beams

This work examines the possibility of using some classical optimization methods (augmented Lagrange multiplier method) or genetic algorithms and neural networks to optimize the cooling regime in hot-rolled complex beam processing. The objective is to optimize the cooling conditions so as to minimize the residual stresses. A detailed example shows that using this stochastic method and neural networks can be efficient. In other words, this work involves finite element analyses of both mechanical and thermodynamic properties of cooling beams, neural networks, genetic algorithms, and optimization to find a cooling regime with minimum residual stresses.     

On the optimization of a rolling-force model for a hot strip finishing line

This paper starts with a profound analysis of rolling-force calculation models of two hot strip rolling product lines. According to classical theory of rolling-force calculation, the structure of rolling-force calculation model is firstly optimized. Then with application of data mining techniques, including clustering sampling approach and optimization algorithm, to the field data from product line, the value of key coefficients of the model is adjusted. By the proposed optimization scheme, a better rolling-force prediction is achieved.     

基于BP神经网络的带钢表面缺陷的识别与分类

运用神经网络处理非线性问题的优势，将其应用于带钢表面缺陷的识别与分类研究。本文采用灰度共生矩阵的特征提取，提出了基于BP神经网络进行缺陷识别与分类的方法，编制了带钢表面缺陷的识别与分类软件。分类测试表明，该软件有较好的识别与分类效果。     

基于神经网络的车辆空气悬架系统阻尼优化

在研究膜式空气弹簧结构参数和受力状况关系的基础上,建立了空气悬架车辆的柔性双质量振动模型,并对模型进行了仿真计算和分析.采用径向基函数神经网络,对空气悬架系统阻尼进行了不同负载工况下的多级优化.结果表明,空气悬架系统阻尼最优值随着载荷的减少而降低,优化阻尼后的车辆,在保证悬架行程的情况下,同时提高了平顺性和轮胎接地性能.     

压电陶瓷驱动器的滑模神经网络控制

由于压电陶瓷驱动器的迟滞非线性严重影响其定位精度,本文提出了一种滑模神经网络控制方法来改善它的性能.用径向基函数神经网络的输出作滑模控制的等价控制量,由迟滞补偿器估计控射器参数误差、外部扰动和近似计算所造成的不确定量对神经网络的输出控制量进行补偿,从而使驱动器系统状态保持在滑模平面上.基于Lyapunov稳定性理论推导了控制器和补偿器的自适应调节律,分析了控制系统的收敛性和稳定性,以可变幅值的低频三角波为参考位移量对控制系统进行了实验测试与分析,结果表明,只采用神经网络控制时的平均定位误差为0.43 μm,最大误差为0.77 μm,而采用滑模控制方法对神经网络控制量进行补偿后,平均定位误差减小为0.27 μm,最大误差减小为0.49μm,定位精度有了显著的提高.     

Similarity of the velocity profile in geometrically distorted flow model

A method is presented to measure the depth-averaged horizontal velocity in distorted physical models, i.e., the depth-averaged horizontal velocity can be equivalent to a point velocity measured at approximately 0.6h below the water surface (h is the water depth). Further a scale for the Chézy coefficient in distorted models is deduced, and the velocity distributions in the vertical in prototype and distorted models are analyzed. By comparing data of the vertical variation of the horizontal velocity measured in a physical distorted model (with distortion ratio of 4.0) with prototype data, inaccuracies via the one-point method, the three-point method and the five-point method are analyzed. It is concluded that it is more accurate to measure the depth-averaged velocity via the three-point method than via the one-point method in the distorted model. What's more, the point velocity at 0.6h below the water surface approximates the depth-averaged velocity in the distorted physical model and the one-point method is proposed to be used in practice.