Motion Track Modeling and Analysis of Belt Spindle under Combined Effects of Bending Moment-Torque-Thermal Deformation

The motion track of belt spindle is important for the radial error of belt spindle. An analytical modeling method for the motion track of belt spindle under the combined effects of bending moment-torque-thermal deformation is proposed in this paper. Three running phases of belt spindle have been analyzed and modeled: the start-up phase with leaping change due to the change in force and bending moment, the accelerating phase with axis deflection, and the constant speed phase with axis regression because of the combined effects of bending moment-torque-thermal deformation. The simulation and test were completed on the belt spindle of SKVM850 machine tool, which illustrates the variation law of the radial error of belt spindle during the whole running phases.     

基于贝叶斯推断的多层软测量建模在丁醇发酵中的应用

针对丁醇生产过程中发酵产物品质参量难以实时测量,现有测量方法精度不高、测量结果受不确定因素影响较大的问题,提出一种基于贝叶斯推断和支持向量回归（Support vector machine regression,SVR）的多层软测量建模方法。首先应用贝叶斯推断计算后验概率、筛选偏置数据,并对偏置数据校准,建立第一层SVR模型;然后利用贝叶斯推断进行二次校准,建立第二层SVR模型,对第一层SVR模型输出进行修正,得到最终预测结果,克服干扰和偏差引起的模型不准确问题。将基于贝叶斯推断的多层支持向量回归（Bi-SVR）预测模型应用于丁醇发酵过程,仿真及实验结果表明,相较于传统SVR预测模型,系统在低干扰的情况下预测精度提高了4.52%,在高干扰时预测精度提高了5.37%。     

Research on control method for machining non-cylinder pin hole of piston

INTRODUCTION With the development of engine power and ro- tational speed, the load of pin hole of piston is more and more heavy. By analyzing the load of common cylindrical pin hole, the load of each part of the pin hole is asymmetric when the engine is working. The stress of inner side is the most and that of the outer side is the least. So, to improve engine performance, people reduce the stress concentration of the inner side of the pin hole by changing the geometric shape and improve …     

数控系统误差补偿新方法研究

1　Introduction　Improvementofmachinetoolaccuracyhasbecomeincreasinglyimportantinmodernmanufacturing .On lywithhighaccuracyofthemachinetoolcanpartswithhighaccuracybeproduced .Therearetwomainapproachestoachieveaccuracyofthemachinetool:erroravoidanceanderrorcompensation .Erroravoid ancetechniquesareusuallyimplementedatthema chinedesignandmanufacturingstages .Becauseofthetechnicalandeconomicallimitations,therearealwayssomeresidualerrors .Therefore ,errorcompensationbecomesanecessaryandpracticals…     

Thermal error compensation for telescopic spindle of CNC machine tool based on SIEMENS 840D system

In this paper, eddy current sensors and thermocouple sensors were employed to measure the thermal field and thermal deformation of a spindle of a telescopic CNC boring-milling machine tool, respectively. A linear regression method was proposed to establish the thermal error model. Furthermore, two compensation methods were implemented based on the SIEMENS 840D system by using the feed shaft of z direction and telescopic spindle respectively. Experimental results showed that the thermal error could be reduced by 73.79% when using the second compensation method, and the thermal error could be eliminated by using the two compensation methods effectively.     

基于BP网络的MEM200快速成形机工艺参数优化

对影响FDM制件翘曲变形的各种工艺参数进行深入分析,结合MEM200快速成形机的特点,确定了影响翘曲变形的五个主要的工艺参数,利用正交试验对主要影响因素进行了实验研究,为神经网络预测模型提供了学习样本。采用贝叶斯正则化算法,通过分析比较确定网络的最佳结构,以提高网络对翘曲变形和成形时间的预测能力。在此基础上,采用优化算法,利用Matlab编制程序实现了工艺参数的优化,经成形验证,表明了该方法能够减少制件的翘曲变形。     

基于AFSA-SVM的船体外板变形预测

船体外板复杂曲面自动化加工一直是船舶制造业研究热点和难点,由于加工过程中船板变形影响因素过多,导致船板加工变形预测一直不够准确快速。鉴于此,将两种复合变量用来表征加工过程中热源对应的众多加工参数,采用人工鱼群算法（AFSA）优化的支持向量机（SVM）预测船板变形。经实验验证,复合参数输入的AFSA-SVM模型预测船体外板水火线加热工艺变形线平均精确度为99.87%,角变形平均精确度为99.53%,且全局最优。将其与传统的PSO-SVM模型对比,不仅精确度有了提高,而且避免了局部极值导致的部分预测结果误差过大情况。     

基于 EMD-PSO-ELM 算法的大坝变形预测研究

基于极限学习机（ELM）构建的大坝变形预测模型易受连接权值、隐含层阈值及隐含层节点数影响,为获取更优的形变预测值,并改善模型预测可靠性,构建一种基于 EMD-PSO-ELM 算法,考虑时效、温度、水位等多因素的大坝变形预测模型。该模型首先从时频分析出发,利用经验模态分解（EMD）将变形时间序列分解成具有不同频率特征的分量|然后利用 PSO-ELM 模型解求形变预测值,重构获得大坝形变趋势。实验结果表明,融合时效、温度、水位等多因素的 EMD-PSO-ELM 大坝变形预测模型残差均方根误差为 1.83mm,平均绝对误差为 1.57mm,平均绝对百分比误差为 1.79%,比 ELM 模型与 EMD-ELM 模型相关误差值更小,在大坝变形预测方面性能更优。     

贝叶斯框架下的自适应质量变量预测模型

针对工业生产过程中的时变性问题,提出贝叶斯网络框架下的自适应质量变量预测建模方法.采用改进的即时学习策略,将数据库分成若干局部数据子集,快速选择与待测样本相似度较高的一组数据作为训练样本,再利用主成分分析对训练样本过程变量进行特征提取,借此作为网络模型输入变量.利用基于改进Figueiredo-Jain算法的EM算法估...     

提高数控机床精度方法研究

通过对数控机床精度的类型特点和影响机床精度原因的分析,了解各类误差补偿方法的应用场合,提出了提高数控机床精度的一些措施。     

热力参数软仪表在电厂中的应用

为了解决电厂中热力参数失效和优化运行的问题,提出了一种基于稀疏高斯过程的软测量建模方法,它基于Bayes在线学习算法,通过构造序列的相关子样本来给出高斯过程的预测输出.通过利用参数化和再生核Hilbert空间范数的投影技巧,得到优化后的参数和后验过程的稀疏高斯逼近.高斯过程的稀疏表达使得基于高斯过程的软仪表能够满足大规模数据集的实时应用需要,所提出的热力参数软仪表对于电厂经济运行有着重要的意义.     

Application of NIR spectroscopy for firmness evaluation of peaches

The use of near infrared (NIR) spectroscopy was proved to be a useful tool for quality analysis of fruits. A bifurcated fiber type NIR spectrometer, with a detection range of 800～2500 nm by lnGaAs detector, was used to evaluate the firmness of peaches. Anisotropy of NIR spectra and firmness of peaches in relation to detecting positions of different parts (including three latitudes and three longitudes) were investigated. Both spectra absorbency and firmness of peach were influenced by longitudes (i,ii, iii) and latitudes (A, B, C). For modeling, two thirds of the samples were used as the calibration set and the remaining one third were used as the validation or prediction set. Partial least square regression (PLSR) models for different longitude and latitude spectra and for the whole fruit show that collecting several NIR spectra from different longitudes and latitudes of a fruit for NIR calibration modeling can improve the modeling performance. In addition, proper spectra pretreatments like scattering correction or derivative also can enhance the modeling performance. The best results obtained in this study were from the holistic model with multiplicative scattering correction (MSC) pretreatment, with correlation coefficient of cross-validation rcv=0.864, root mean square error of cross-validation RMSECV=6.71 N, correlation coefficient of calibration r=0.948, root mean square error of cali-bration RMSEC=4.21 N and root mean square error of prediction RMSEP=5.42 N. The results of this study are useful for further research and application that when applying NIR spectroscopy for objectives with anisotropic differences, spectra and quality indices are necessarily measured from several parts of each object to improve the modeling performance.     

基于支持向量机的汽车故障预测贝叶斯网络推理系统研究

相较于传统的汽车故障诊断,汽车故障预测通过对汽车运行状态数据的分析,实现对汽车未来发生故障的可能性和故障类型进行预测和分析,充分保证汽车行驶的安全。通过对汽车故障类型和特点的深入研究,在已有的故障预测技术的基础上,设计了一种数据驱动的故障预测推理系统。该系统利用支持向量机完成对汽车不同类型故障的分类,结合贝叶斯推理网络,实现对当前汽车状态的综合分析,得出汽车在未来发生不同类型故障的概率,进而完成对汽车故障的预测。最后,编写软件实现汽车故障预测系统。     

基于变周期三角函数—灰色GM(1，1)模型的设备故障间隔时间预测

针对设备故障预测模型难以精确建立的特点,为提高故障间隔时间预测的精确度,提出了变周期三角函数-灰色模型GM(1,1)的预测方法。该方法在三角函数-灰色模型GM(1,1)基础上,建立了变周期三角函数—灰色GM(1,1)的组合模型,实现了对设备故障间隔时间的预测;并将预测结果与三角函数-灰色模型GM(1,1)进行对比,结果表明,采用变周期三角函数-灰色模型GM(1,1)对故障间隔时间进行预测,其预测结果的相对误差由24.16%降到3.24%,提高了预测结果的精度。     

平面磨削工件的热变形及加工误差的实用计算方法

The paper submits a method to calculate thermal deformation and manufacture error in surface grinding.The author established a simplified temperature field model.and derived the thermal deformaiton of the ground workpiece,It is found that there exists not only a upwarp thermal deformation,but also a parallel expansion thermal deformation.A upwarp thermal deformation causes a concave shape error on the profile of the workpiece,and a parallel expansion thermal deformation causes a dimension error in height.The calculations of examples are given and compared with presented experiment data.     

超大尺寸超精密数控磨床构型分析

为确定用于超大口径（>=1500mm）光学非球面成形加工的超精密数控磨床的构型设计方案,本文开展了2种不同构型龙门机床的空间运动误差建模及比较研究。基于多体系统理论,通过建立2种构型龙门机床的拓扑结构图,推导二者的空间运动误差模型;并基于该模型,采用仿真手段比较了二者用于超大尺寸光学非球面加工过程的砂轮中心点误差变化规律;仿真结果表明,在加工大口径非球面元件时,工作台移动式的机床构型具有相对较小的空间运动误差,但两种构型龙门机床的误差十分接近。在综合考虑二者在大口径非球面元件加工过程中的其他特点后,确定了该超精密磨削机床构型为工作台固定的桥式龙门结构。     

基于自适应无味卡尔曼滤波的机床选点温升快速辨识方法研究

研究目的：为了缩短机床温升试验时间,提出一种机床热特性快速辨识方法,利用较短时间的温度采样数据即可准确预测出完整的温升曲线,进而获得热平衡时间及稳态温度等热特性参数。创新要点：提出了基于自适应无味卡尔曼滤波的机床选点温升快速辨识方法,其中最短辨识时间判据可以有效解决如何寻找准确辨识热特性参数的最短采样时间问题,而自适应无味卡尔曼滤波则可以实时调整参数,防止外界因素对辨识的干扰。研究方法：由于无味卡尔曼滤波在非线性状态预测和参数辨识上具有优势,所以本文将无味卡尔曼滤波算法应用到机床选点温升辨识上。为了防止辨识过程中的发散退化等问题,将无味卡尔曼滤波发展为自适应无味卡尔曼滤波（图1）。在快速辨识方法上提出了最短辨识时间判据（图2）。文章中又将此算法应用到实际的立式加工中心温升辨识上,证明了该算法的可行性及有效性（图5和6）。最后又将带有自适应调整过程的无味卡尔曼滤波算法和不带调整过程的算法做了对比,显示了自适应调整过程对辨识算法的重要性（图6和11）。重要结论：基于自适应无味卡尔曼滤波的机床选点温升快速辨识方法可以准确快速地辨识出温升曲线,获取热特性参数,将原来394 min的热平衡试验时间缩短,只需28 min即可得到温升变化情况。     

MJ-50数控车床主轴回转精度的影响因素分析

以MJ-50型数控车床为例,通过分析主轴瞬时回转轴线的轴向窜动、径向跳动和纯角度摆动,以此确定主轴回转精度的评定指标;并对影响主轴精度的几种主要因素进行了深入探讨,为确定数控机床加工精度范围提供了一种新方法。     

方形锂离子电池热行为集总建模仿真与实验

针对锂离子电池电化学产热机理复杂、热分析建模参数多等问题,引入集总模型思想,建立单体方形电池少参化产热模型。通过大量实验、数据分析,辩识电池热物性参数,获得基于真实物理参数的热行为仿真集总模型。对比单体电池在不同放电倍率下温度场的实验数据和仿真结果发现：在0.2、0.5、1 C恒流放电工况下,集总模型对电池表面温度的预测误差均低于0.55℃,误差率小于1.94%,达到实际工程预测精度要求。该研究结果可为商业化锂离子电池热行为分析提供参考。     

加工中心在线检测系统误差补偿技术研究

基于多体理论，建立了加工中心在线检测系统的几何误差与热误差综合模型，对测头误差处理技术进行了研究．利用该模型可以同时对测头误差、机床厂一何误差与热误差进行补偿，实验表明，此方法有效地提高了在线检测精度。