Training effect on damping capacity in Fe–20 mass% Mn binary alloy

Two-dimensional finite element analysis together with stream function and neural network models are employed to determine thermo-mechanical behavior during hot strip rolling of AA5083. An appropriate velocity field and stream function is first determined using the rule of volume constancy and upper bound theorem and then temperature field within the metal is predicted by means of a two-dimensional conduction–convection model. In order to consider the effect of flow stress and its dependence on temperature, strain and strain rate, a neural network model is also employed in the analysis. Based on the performed tensile tests, two different neural network models are constructed one for smooth yielding and the other one for the serrated flow. Then, the ANN models are coupled with the thermo-mechanical analysis. In the next step, by combination of the predicted temperature, strain and strain rates fields and the experimental data achieved from the tensile tests, the occurence of dynamic strain ageing during hot rolling is predicted. The model predictions are then compared with the experimental data and good agreement is observed between the two sets of results that shows the validity of the proposed model.     

神经网络模型在SiC涂层制备中的应用

材料表面抗氧化涂层的质量是限制碳／碳复合材料作为高温结构材料使用的关键．本文运用人工神经网络技术建立了CVD－SiC涂层制备工艺的过程模型,以解决该过程影响因素众多、相互作用关系复杂、难以对制备过程进行有效的预测和控制的问题．研究结果表明：所建立的神经网络模型,可以比较准确和全面地反映各工艺因素对SiC-CVD过程的影响大小及内在规律;模型对工艺参数与沉积速率之间关系的预测与实验结果相吻合;证实了将人工神经网络模型应用于抗氧化涂层的制备过程的控制和工艺优化是有效和可行的．     

An exploratory study of a neural network approach for reliability data analysis

The results of this paper show that neural networks could be a very promising tool for reliability data analysis. Identifying the underlying distribution of a set of failure data and estimating its distribution parameters are necessary in reliability engineering studies. In general, either a chi-square or a non-parametric goodness-of-fit test is used in the distribution identification process which includes the pattern interpretation of the failure data histograms. However, those procedures can guarantee neither an accurate distribution identification nor a robust parameter estimation when small data samples are available. Basically, the graphical approach of distribution fitting is a pattern recognition problem and parameter estimation is a classification problem where neural networks have been proved to be a suitable tool. This paper presents an exploratory study of a neural network approach, validated by simulated experiments, for analysing small-sample reliability data. A counter-propagation network is used in classifying normal, uniform, exponential and Weibull distributions. A back-propagation network is used in the parameter estimation of a two-parameter Weibull distribution.     

Modeling constitutive relationship of Ti40 alloy using artificial neural network

Constitutive relationship equation reflects the highly non-linear relationship of flow stress as function of strain, strain rate and temperature. It is a necessary mathematical model that describes basic information of materials deformation and finite element simulation. In this paper, based on the experimental data obtained from Gleeble-1500 Thermal Simulator, the constitutive relationship model for Ti40 alloy has been developed using back propagation (BP) neural network. The predicted flow stress values were compared with the experimental values. It was found that the absolute relative error between predicted and experimental data is less than 8.0%, which shows that predicted flow stress by artificial neural network (ANN) model is in good agreement with experimental results. Moreover, the ANN model could describe the whole deforming process better, indicating that the present model can provide a convenient and effective way to establish the constitutive relationship for Ti40 alloy.     

On the pyrolysis kinetics of scrap automotive tires

Pyrolysis kinetics of scrap tires of passenger car and truck have been investigated thermogravimetrically under heating rates of 5, 10, 20 and 30K/min and temperature range 373-1273K in nitrogen. The results show that the initial reaction temperatures are 482-521K for the tire of passenger car and 458-511K for truck tire. Both tires exhibit similar behaviors that the initial reaction temperature decreases, but reaction range and reaction rate increase when heating rate is increased. The overall rate equation for each tire can be modeled satisfactorily by a simple one equation from which the kinetic parameters such as the activation energy (E), the pre-exponential factor (A), and the reaction order (n) of unreacted material based on Arrhenius form are determined using Friedman's method. The results show that two tires behave similarly and the average kinetic parameters of two tires are E = 147.95 +/- 0.21kJ/mol, A = (6.295 +/- 1.275)x10(10)min(-1), and n = 1.81 +/- 0.18. The predicted rate equations compare fairly well with the measured data.     

Application of Statistical Techniques and Neural Networks in Condition‐Based Maintenance

In this paper, we have evaluated five prediction approaches from two disciplines for condition‐based maintenance. It also includes a case study for vehicle tire pressure monitoring as an example application. Main focus of this paper is on two widely used areas in prediction: (i) statistics, (ii) neural networks. It is well known that these two areas have wide applications in forecasting. Statistical and neural network techniques are very powerful for predicting the future states based on current and previous states of the system or subsystem. Application of ARAR and Holt‐Winters (HW) in CBM has been presented from the statistics point of view. On the other hand, application of focused time delay, linear predictor, and backpropagation neural network has also been presented to prove the robustness of statistical approaches. Paper presents detailed comparative simulation study to show the suitability and feasibility of all the techniques. We assumed that the sensors are directly mounted on tires externally and report the current tire pressure to control or analysis. The control unit performs tire pressure analysis and reports the decision to operator or intended group about current pressure as well as the impending pressure conditions. Finally, investigation ends with conclusion that HW is best suited among these five approaches for tire pressure prediction and could be useful to design a CBM application for any system. Copyright © 2012 John Wiley & Sons, Ltd.     

基于非线性黏弹性本构模型的轮胎滚动和生热

轮胎的滚动阻力和自生热是造成轮胎失效的原因之一,轮胎内部能量的产生主要取决于轮胎中橡胶材料的黏弹性能量耗散。文中基于超弹性模型和并行流变模型(PRF)描述了橡胶材料的非线性黏弹性响应特征,提出了一种预测实心轮胎温度分布和滚动阻力的方法。首先将线性黏弹性Prony级数转化为PRF模型的初始参数,并利用Isight软件根据多应变工况加载得到的应力松弛测试数据来校准材料参数,然后采用显式-热力耦合分析方法分析基于Prony级数和PRF模型的实心轮胎滚动过程的差异。结果表明,Prony级数无法描述橡胶材料的非线性行为,在显式动力学下计算的轮胎生热结果为0;PRF模型可以表征橡胶材料的非线性行为,并且计算的轮胎模型在0.3s内温度上升了0.14℃。     

汽车轮胎动特性的理论评价

在汽车动力学中，汽车轮胎是最重要的零件之一，从汽车轮胎设计观点来看，轮胎动特性的模式化是非常重要的。因为这些轮胎特性影响着汽车操纵稳定性和操纵灵活性，提供了一种轮胎动特性的模型，它是在Fiala提出的稳定状态轮胎模型和Pacejka提出的组合型基模础上建立的。被提出的模型是用轮胎物理参数和两个动特性指数来表示的。     

Artificial Neural Network for Creep Behaviour Predictions of a Parallel‐lay Aramid Rope Under Varying Stresses

Abstract: An improvement of the creep behaviour prediction of parallel‐lay aramid ropes under varying stresses is the scope of the following study in which application of artificial neural networks (ANNs) for the prediction of creep under varying stresses is presented. This qualitatively different approach assumes that the ANN can be trained to simulate time‐dependent response of the rope in the given load (stress) programme and time interval. The classic rheological constitutive equations are not needed in this case, because ANN acts as a constitutive operator trained by stresses and the corresponding creep strains from experimental data. Carried numerical experiments were divided into the following three parts: (i) searching the best ANN for a creep behaviour approximation under varying stresses, (ii) investigating the best topology of the selected neural network and (iii) investigating the best results for the creep function identification. Comparison between the experimentally observed creep strains of the parallel‐lay aramid rope under varying stresses, predicted creep strains when the linear creep constitutive equation is applied and predicted creep strains when the obtained Jordan neural network with the 3‐10‐1 topology is used confirmed that the Jordan neural network developed achieved less than half mean square error beside the existing creep constitutive analytical approach.     

Artificial neural network‐based analysis of effective crack model in concrete fracture

Many non‐linear fracture models have been proposed by design codes and investigators to determine fracture parameters of cement‐based materials. To characterise failure of concrete structures, the effective crack model (ECM) needs two fracture parameters: the effective crack length a_e and the critical stress intensity factor . Nevertheless, ECM requires a closed‐loop testing system and the calculation of a_e needs considerable computational effort. For this reason, ECM is simulated with an artificial neural network (ANN) in this study. The main benefit of using an ANN approach is that the network is built directly on experimental data by using the self‐organizing capabilities of the ANN. The presented fracture model was developed by utilising 464 noisy test data taken from the literature, which were obtained via different test methods in different laboratories. The results of an ANN‐based ECM look viable and very promising.