Prediction and simulation errors in parameter estimation for nonlinear systems

This article compares the pros and cons of using prediction error and simulation error to define cost functions for parameter estimation in the context of nonlinear system identification. To avoid being influenced by estimators of the least squares family (e.g. prediction error methods), and in order to be able to solve non-convex optimisation problems (e.g. minimisation of some norm of the free-run simulation error), evolutionary algorithms were used. Simulated examples which include polynomial, rational and neural network models are discussed. Our results—obtained using different model classes—show that, in general the use of simulation error is preferable to prediction error. An interesting exception to this rule seems to be the equation error case when the model structure includes the true model. In the case of error-in-variables, although parameter estimation is biased in both cases, the algorithm based on simulation error is more robust.     

一种结构可靠性的模拟数值计算方法

常用结构可靠性计算的方法包括有蒙特卡洛模拟法,响应面法等。大多数方法在计算时必须采用有限元理论,并进行基于Monte carlo抽样仿真,以研究多种不确定因素的影响下结构的可靠性。因此文中提出了一种基于kriging理论建模和遗传算法求解的模拟数值可靠性分析方法,通过采用重要抽样方法以减少抽样仿真次数,并利用kriging模型的高效预测建立结构的状态功能函数,在此基础上,应用遗传算法计算结构可靠性指标。将此方法用于典型框架结构可靠度分析的实例,表明了该方法的有效性和较高的计算效率。     

基于小波去噪和LSTM模型的短时交通流预测

针对现有的部分交通流预测模型仅面向单一路段进行,模型输入数据未预处理的问题,采用启发式阈值算法对小波分解后的原始交通流数据进行去噪处理,通过对路网内各路段交通流数据相关性系数计算,构造出路网交通流数据压缩矩阵。数据去噪将数据对模型的干扰降到最低,同时路网数据相关性分析又使预测在路网层面上进行了考量。利用长短时记忆(LSTM)网络在时序数据处理方面的优势,将压缩矩阵输入构造好的LSTM模型进行短时交通流预测。利用去噪处理数据和原始数据分别训练LSTM-1和LSTM-2模型,通过仿真实验,设置不同预测时间将本文提出的预测方法和其他4种模型对比,验证了相较于其他4种模型预测的准确率平均可提升10.278%,预测的准确率达到了95.58%,说明这是一种高效率的短时交通流预测方法。     

车辆串联混合传动中二次调节系统的节能制动

建立了采用二次调节静液传动技术的车辆串联混合传动系统的数学模型，分析了以恒扭矩制动和恒功率制动回收车辆惯性能的特点，以满载的天津三峰牌TJ6481A型车辆作为对象进行了恒扭矩、恒功率制动的仿真研究，并在模拟实验台上进行了模拟实验，从车辆的制动可靠性和乘坐舒适性两方面对这两种制动方法的可应用性进行了分析。     

基于Archard模型的机床导轨磨损模型及有限元分析

机床导轨的磨损是导致机床精度降低或丧失的重要因素。针对机床导轨磨损深度的预测问题,提出了一种基于Archard模型与有限元模拟试验的机床导轨磨损分析方法。该方法基于修正的Archard模型形成一种磨损深度的计算公式,并采用一种离散化计算方法进行求解;结合ANSYS软件,提出机床导轨磨损有限元分析模拟的方法及流程;最后,以车床滑动导轨为例,用有限元试验方法对机床导轨的磨损过程进行模拟分析,得到磨损深度关于磨损次数的计算公式,其结果表明:机床导轨磨损深度的仿真模拟值与实际值具有一致性。     

基于多种信号分解的台风风速多步预测

针对台风风速多步预测中用到的信号分解方法开展对比研究。首先,列举8种典型信号分解方法的特点;其次,基于不同信号分解方法建立经粒子群算法(particle swarm optimization,简称PSO)优化的最小二乘支持向量机(least squares support vector machine,简称LSSVM)预测模型;最后,采用某大跨桥梁主塔位置和沿海某高层建筑楼顶处的两组台风实测风速序列进行多步提前预测研究。对两组试验的预测结果进行分析,发现基于变分模态分解(variational mode decomposition,简称VMD)的PSO-LSSVM模型具有最佳预测效果。     

自升式海洋平台固桩架结构的疲劳分析

针对海洋平台结构,传统的疲劳分析方法是采用应力校核．按照现有材料试样疲劳特性,运用经验公式,在大量简化基础上进行计算的方法,通常得到一个可行的、但不是最简便的疲劳分析方案。文中以某300英尺自升式海洋平台的固桩架为研究对象,在nSoft疲劳分析理论的基础上,运用计算机模拟试验法,估计固桩架构件的S-N曲线;在固桩架工作的时间历程里,确定其载荷谱块,进行寿命预测,为设计人员对类似结构的设计,估计其安全性提供了一定的参考价值。     

膜片稳压溢流阀的瞬态数值模拟方法研究

论述了计算流体动力学(CFD)分析中生成动态网格的主要技术,并应用在稳压膜片溢流阀瞬态CFD动态网格的划分和计算,保证了模拟的不失真,从而较好地解决了由于阀瓣运动所导致计算区域瞬时变化的问题。通过与实测值的比较,验证了CFD数值模拟在分析和预测溢流阀特性上的可行性。     

塑料注射模具的并行设计方法

塑料制品注射成型属于一次成型工艺 ,传统的串行模具设计和制造方式 ,造成了对设计经验的过分依赖和模具返修率的提高 ,本文举例介绍了基于流动模拟技术的注射模并行设计方法 ,说明了 CAE对模具结构设计和相关注塑工艺的指导作用。     

On the prediction of side-wall wrinkling in sheet metal forming processes

Prediction and prevention of side-wall wrinkling are extremely important in the design of tooling and process parameters in sheet metal forming processes. The prediction methods can be broadly divided into two categories: an analytical approach and a numerical simulation using finite element method (FEM). In this paper, a modified energy approach utilizing energy equality and the effective dimensions of the region undergoing circumferential compression is proposed based on simplified flat or curved sheet models with approximate boundary conditions. The analytical model calculates the critical buckling stress as a function of material properties, geometry parameters and current in-plane stress ratio. Meanwhile, the sensitivities of various input parameters and integration methods of FEM models on the prediction of wrinkling phenomena are investigated. To validate our proposed method and to illustrate the sensitivity issue in the FEM simulation, comparisons with experimental results of the Yoshida buckling test, aluminum square cup forming and aluminum conical cup forming are presented. The results demonstrate excellent agreements between the proposed method and experiments. Our model provides a reliable and effective predictor for the onset of side-wall wrinkling in sheet metal forming processes.     

利用Kriging预测模型的EMD端点延拓

结合端点效应的产生机理和关于端点效应的现有研究成果,提出了一种基于Kriging预测模型的抑制EMD端点效应的新方法.通过计算信号及其包络线的最优线性无偏预测,将信号的上、下包络线进行延拓,从而最大化地逼近原始信号两端点,并将未延拓的分解结果、基于镜像延拓法得到的结果和基于Kriging预测延拓法得到的结果进行对比分析.仿真算例和试验结果表明,基于Kriging预测模型的延拓方法抑制EMD端点效应的效果最优,能够精确反映信号特征,有利于准确提取结构的模态参数,提高运算效率.     

六自由度机器人在焊缝焊接应用中的路径规划

常用的机器人轨迹规划有关节空间法和直角坐标空间法,本文分析了这2种轨迹规划的特点。提出了分段二次Lagrange插值路径规划法,并进行了计算机仿真。仿真结果表明,采用分段二次Lagrange插值的方法能使机器人末端的运动轨迹平滑且能很好地逼近于给定的焊缝曲线。     

考虑轴承间隙时裂纹转子的故障状态及仿真分析

基于所建立的开闭裂纹转子的动力学模型,并考虑到轴承间隙这一非线性因素,对转子在不同激励参数下的裂纹开闭和间隙力存在条件进行了预测,并对转子运行轨道和频谱响应进行了动态仿真分析,结果表明,对裂纹开闭和间隙力存在条件的预测是正确的,同时转子出现裂纹和间隙力时显示出特殊的动力学特性。     

机电一体化系统建模技术与仿真软件的研究与分析

总结了国际上目前研究的几种可以实现自动向数学模型转化的机电一体化系统理想的物理模型建立的方法，即键合图方法、面向对象的方法、方块图方法、系统图方法和混合Petri网方法。分析研究了由Lancaster大学EDC(Engineering Design Center)中心的研究者们开发的计算机辅助机电一体化系统概念设计的建模与仿真软件Schemebuilder，研究了用键合图方法建模的仿真软件20-sim和使用面向对象方法建模的仿真软件Dymola，以及使用方块图建模的控制系统仿真软件Matlab和机械机构系统仿真软件Adams。最后总结了机电一体化建模与仿真的发展趋势，即机电相结合的机电一体化系统仿真软件为机电一体化系统理想的建模和仿真环境。     

Multiple optimized online support vector regression for adaptive time series prediction

Data-driven prognostics based on sensor or historical test data have become appropriate prediction means in prognostics and health management (PHM) application. However, most traditional data-driven prognostics methods are off-line which would be seriously limited in many PHM systems needed on-line predicting or real-time processing. Furthermore, even in some on-line prediction algorithms such as Online Support Vector Regression (Online SVR) and Incremental learning algorithm, there are conflicts and trade-offs between prediction efficiency and accuracy. Therefore, in different PHM applications, prognostics algorithms should be on-line, flexible and adaptive to balance the prediction efficiency and accuracy. An on-line adaptive data-driven prognostics strategy is proposed with five various optimized on-line prediction algorithms based on Online SVR. These five algorithms are improved with kernel combination and sample reduction to realize higher precision and efficiency. These algorithms can achieve more accurate results by data pre-processing and model optimization, moreover, faster operating speed and lower computational complexity can be obtained by optimization of training process with on-line data reduction. With these different improved Online SVR methods, varies of prediction with different precision and efficiency demands could be fulfilled by an adaptive strategy. To evaluate the proposed prognostics strategy, we have executed simulation experiments with Tennessee Eastman (TE) process. In addition, the prediction strategies are also applied and evaluated by traffic mobile communication data from China Mobile Communications Corporation Heilongjiang Co., Ltd. Experiments and test results prove its effectiveness and confirm that the algorithms can be effectively applied to the on-line status prediction with increased performance in both precision and efficiency.     

Modeling of the centerless infeed (plunge) grinding process

A computer simulation method for investigating the form generation mechanism in the centerless infeed (plunge) grinding process is described. For a 3-D simulation model of form generation, contact points are assumed to be on least squares contact lines at the grinding wheel, regulating wheel, and work-rest blade. Using force and deflection analyses, the validity of this assumption is shown. Based on the 2-D simulation model developed in the previous work and the least squares contact line assumption, a 3-D model is presented. To validate this model, simulation results were compared with the experimental works. The experiments and computer simulations were carried out using three types of cylindrical workpiece shapes with varying flat length. The experimental results agree well with the simulation. It can be seen that the effect of flat end propagated to the opposite end through workpiece reorientation.     

基于多嵌入维数的时用水量LSSVM组合预测

为解决供水系统调度所需混沌时用水量高精度预测等问题,将最小二乘支持向量机(LSSVM)组合预测模型应用到城市时用水量预测中。在分析不同嵌入维数和预测方法对模型预测精度影响程度的基础上,提出了基于多嵌入维数的LSSVM组合预测模型。采用互信息法和G-P方法求取多个嵌入维数,并建立了不同相空间模型,通过LSSVM算法对上述多个预测模型进行了组合预测,既综合了各不同嵌入维数各预测方法下的信息,又对单一模型下的预测偏差进行了融合,以有效地提高预测精度;最后在某地进行了时用水量序列的仿真实验。研究结果表明,该模型预测精度平均误差小于2%,明显优于各单一模型的预测结果,证实了该组合模型的有效性和实用性。     

Prediction of warpage in plastic injection molding based on design of experiments

In terms of injection processing parameters, a mathematical model for prediction of warpage was formulated based on design of experiments (DOE). First, the five most influential parameters were screened by using fractional factorial design (FFD): melt temperature, coolant temperature, injection time, V/P switch over and mold temperature. Second, considering the other four principal processing parameters except the melt temperature, the predicting mathematical model was founded by using central composite design (CCD) of experiments and FE simulation. Finally, the results of statistical analysis were collected from software Moldflow. The results suggested that the mathematical model can be used to predict warpage with adequate accuracy. Hence, it indicated that corrective and iterative design steps can be initiated and implemented for better quality of products without resorting to physical trials in plastics injection mold by using this predicting mathematical model.     

Springback prediction and optimization of variable stretch force trajectory in three-dimensional stretch bending process

Most of the existing studies use constant force to reduce springback while researching stretch force. However, variable stretch force can reduce springback more efficiently. The current research on springback prediction in stretch bending forming mainly focuses on artificial neural networks combined with the finite element simulation. There is a lack of springback prediction by support vector regression(SVR). In this paper, SVR is applied to predict springback in the three-dimensional stretch bending forming process, and variable stretch force trajectory is optimized. Six parameters of variable stretch force trajectory are chosen as the input parameters of the SVR model. Sixty experiments generated by design of experiments(DOE) are carried out to train and test the SVR model. The experimental results confirm that the accuracy of the SVR model is higher than that of artificial neural networks. Based on this model, an optimization algorithm of variable stretch force trajectory using particle swarm optimization(PSO) is proposed. The springback amount is used as the objective function. Changes of local thickness are applied as the criterion of forming constraints. The objection and constraints are formulated by response surface models. The precision of response surface models is examined. Six different stretch force trajectories are employed to certify springback reduction in the optimum stretch force trajectory, which can efficiently reduce springback. This research proposes a new method of springback prediction using SVR and optimizes variable stretch force trajectory to reduce springback.     

The relation between the friction contact performance and the microgeometry of contacting surfaces

It is demonstrated how computer simulation can enable the assessment of the effect of microgeometrical parameters on some friction contact features. The findings of model experiments demonstrate how the distribution of the heights and radii of asperities affects the contact deformation and real contact area. The contact model reveals how the roughness parameters influence the friction coefficient and the contact conductivity. It is illustrated that computer simulation can be applied to the rating of the effect of temperature on the contact conductivity.