2Cr13不锈钢高速铣削表面粗糙度预测模型研究

针对汽轮机叶片常用钢2Cr13不锈钢在切削加工中表面质量存在的问题,对高速铣削条件下2Cr13不锈钢表面粗糙度预测模型进行了研究。将最小二乘支持向量机原理应用到高速铣削2Cr13不锈钢的表面粗糙度预测建模中。得出的模型能方便地预测铣削参数对表面粗糙度的影响,并能利用有限的试验数据得出整个工作范围内的表面粗糙度预测值。经试验验证,应用最小二乘支持向量机原理建立的粗糙度预测模型回归预测精度高。基于最小二乘支持向量机原理建模方法适合于表面粗糙度预测。     

铣削加工粗糙度的智能预测方法

提出了一种基于最小二乘支持向量机的铣削加工表面粗糙度智能预测方法.首先进行了铣削工艺参数对工件表面粗糙度影响的正交实验,再通过对主轴转速、进给速率和切削深度三因素,以及各因素之间交互三水平实验的数据分析,找出了铣削工艺参数对工件表面粗糙度影响的一些规律.利用最小二乘支持向量机算法建立了铣削预测模型,通过该模型能在有限实验基础上利用工艺参数方便地得到粗糙度预测值.实际预测表明,在相同情况下,该模型构造速度比反向传播神经网络建模预测方法高2个～3个数量级,预测精度高10倍左右.     

基于LS-SVM的汽轮机叶片高速铣削表面质量预测建模

针对汽轮机叶片高速铣削加工中存在的表面质量不易控制问题,借助最小二乘支持向量机原理,建立了被加工不锈钢叶片表面的粗糙度预测模型。实验结果表明,该模型能方便地预测切削速度、主轴转速、进给量、铣削宽度等铣削参数对铣削加工工件表面粗糙度的影响,并能利用有限的实验数据得出整个工作范围内的表面粗糙度预测值,模型适合于表面粗糙度预测,回归预测精度高。     

航空铝合金三维端铣表面粗糙度的LS-SVM控制研究

为提高加工工件的表面质量,需要有效控制加工工件表面粗糙度,因此有必要建立精度高、泛化能力强的表面粗糙度预测模型。首先基于具有位错动力学物理基础的Z-A材料本构模型,建立航空铝合金7050材料的三维端面铣削有限元仿真模型,并设计正交试验验证有限元模型的可靠性;其次建立最小二乘支持向量机(LS-SVM)预测模型,以仿真所提供的样本数据为输入,拟合铣削参数与表面粗糙度的复杂非线性关系,实现了表面粗糙度的预测,结果表明LS-SVM模型预测的相对误差不超过6%;最后基于LS-SVM表面粗糙度预测模型得出各铣削参数对表面粗糙度的影响,为生产实际提供指导。     

最小二乘支持向量机在U71Mn高锰钢表面粗糙度预测模型中的运用

获取了U71Mn高锰钢在特定主轴转速n、进给量f、铣削深度ap、铣削宽度ae加工条件下的表面粗糙度Ra的原始数据。基于留出法原则将原始数据依次随机分为两组,一组为训练集用于训练U71Mn高锰钢的预测模型;另一组数据为验证集用于验证模型,并且通过机器学习性能评价指标来确定模型的最终预测精确率。通过实际建模对比发现最小二乘支持向量机预测模型其拟合以及预测精度明显高于传统多元线性回归模型。最小二乘支持向量机(LSSVM)通过对原支持向量机算法(SVM)进行了算法改进,在算法中把原求解Lagrange乘子α不等式约束的二次规划(QP)问题,转化为等式约束即求解线性方程组,显著减少了计算机运算的时间复杂度。并且通过寻求结构化风险最小提高了学习机的泛化能力,在观测样本数量较小的情况下,容易实现经验风险和置信范围的最小化,使模型对未知样本有良好的鲁棒性与预测精度。     

基于最小二乘支持向量机的外圆磨削表面粗糙度预测系统

为解决磨削加工中影响因素多,难以实现自动化加工的困难,对磨削系统的表面粗糙度预测系统进行了研究。在分析目前常用预测方法的基础上,建立了基于最小二乘支持向量机的外圆纵向磨削表面粗糙度预测模型。该模型采用等式约束,把原来求解一个二次规划问题转化成求解一个线性方程组,方法简单且有效。比较实验显示,该方法响应时间快、测量精度高,测量精度误差比BP神经网络预测方法小4%,比进化神经网络(BP+GA)预测方法小1.3%,所提供的预测方法可以实现对工件表面粗糙度的在线预测。将其应用于外圆纵向磨削智能系统中,实时计算预测值与给定粗糙度的差值,引导磨削专家系统修正磨削参数,实现智能控制,取得了较好的效果。     

基于最小二乘支持向量机的疲劳裂纹扩展预测

根据腐蚀疲劳裂纹在扩展过程中受到多种环境因素影响,裂纹扩展预测难精确的特点,本文提出了基于遗传算法参数优化的最小二乘支持向量机方法来预测结构腐蚀疲劳裂纹扩展。该算法采用遗传算法优化最小二乘支持向量机的模型参数,从而避免了算法陷入局部最优解,实现了精确度高、泛化能力强的裂纹扩展预测模型。最后通过对已有文献的某试件裂纹扩展的实验数据进行建模分析。结果表明:基于遗传算法的最小二乘支持向量机预测方法优于神经网络算法、蚁群算法,预测误差较小,具有很好的预测能力。     

切削加工表面粗糙度组合预测模型研究

加工过程产生的粗糙度数据序列会包含多种特征,而单一的预测模型不能同时捕捉多种数据特征,难以提高预测精度。因此,从加工过程中粗糙度数据特征的复杂性出发,提出了一种基于支持向量机(SVM)和BP神经网络算法(BP)的组合预测模型,来同时捕捉数据的线性特征和非线性特征;在组合预测过程中为充分发挥两种预测算法的最佳性能,采用粒子群优化算法(PSO)对支持向量机的参数和BP神经网络中的权值进行优化。通过蠕墨铸铁的铣削实验,实现不同切削用量下的表面粗糙度精准预测,并与PSO-SVM、PSO-BP算法以及切削加工表面粗糙度理论模型进行对比,验证了该组合模型的优越性。     

基于IAGA-LSSVM的切削加工表面粗糙度的智能预测

提出了一种基于改进自适应遗传算法与最小二乘支持向量机(IAGA-LSSVM)的切削加工表面粗糙度的智能预测方法。通过设定LS-SVM模型主要参数的取值范围,采用IAGA进行寻优,提高了LS-SVM预测模型的精度。最后采用平均相对预测误差作为检验指标,比较了多元线性回归模型、BP神经网络模型、AGA-LSSVM模型及IAGA-LSSVM模型对表面粗糙度的预测能力。结果表明:IAGA-LSSVM预测模型的建模时间更短,平均相对预测误差更小,对切削加工表面粗糙度的预测具有一定的参考意义。     

基于最小二乘支持矢量机的成形磨削表面粗糙度预测及磨削用量优化设计

在分析和比较目前常用的预测方法基础上,提出一种基于最小二乘支持矢量机的成形磨削表面粗糙度预测方法。一方面,该方法能较好地解决小样本学习问题,避免人工神经网络等智能方法在对粗糙度进行预测时所表现出来的过学习、泛化能力弱等缺点;另一方面,用等式约束代替传统支持矢量机的不等式约束,减小了模型的复杂度,加快了求解速度。试验表明,该模型具有预测精度高、速度快、容易实现等优点,适合对磨削表面粗糙度的预测。在成功建立预测模型的基础上,还提出磨削参数优化设计的可行性方案,建立表面粗糙度与磨削用量之间的关系图,对于优化设计磨削用量、提高加工零件表面质量具有一定的指导意义。     

侧铣加工刀具回转轮廓误差预测技术研究

在侧铣加工中，刀具磨损和变形引起的刀具回转轮廓误差在实际加工前难以准确预测。提出一种工件形状刀具轮廓映射的辨识试验方法来获取加工过程刀具回转轮廓误差，并通过多因素正交试验获取了不同工况下刀具回转轮廓误差数据库。基于误差数据，采用最小二乘支持向量机（LS-SVM）技术建立了刀具回转轮廓误差预测模型。运用遗传算法优化对所提模型有重要影响的核函数参数和错误惩罚因子, 建立了基于遗传算法优化的最小二乘支持向量机(GA-LS-SVM)模型，并与未经遗传算法优化的LS-SVM模型进行了对比，试验结果表明，GA-LS-SVM预测模型能更好地适用于刀具回转轮廓误差预测。     

Effect of the process parameters on the three-dimensional shape of molten pool during full-penetration laser welding process

Slow tool servo (STS) turning is superior in machining precision and in complicated surface. However, STS turning is a complex process in which many variables can affect the desired results. This paper focuses on surface roughness prediction in lenses STS turning. An exponential model, based on the five main cutting parameters including tool nose radius, feed rate, depth of cut, C-axis speed, and discretization angle, for surface roughness prediction of lenses is developed by means of orthogonal experiment regression analysis. Meanwhile, a prediction model of surface roughness based on least squares support vector machines (LS-SVM) with radial basis function is constructed. Orthogonal experiment swatches are studied, and chaotic particle swarm optimization and leave-one-out cross-validation are applied to determine the model parameters. The comparison of LS-SVM model and exponential model is also carried out. Predictive LS-SVM model is found to be capable of better predictions for surface roughness and has absolute fraction of variance R² of 0.99887, the mean absolute percent error e_M of 8.96 %, and the root mean square error e_R of 10.68 %. The experimental results and prediction of LS-SVM model show that effects of tool nose radius and feed rate are more significant than that of depth of cut on surface roughness of lenses turning.     

Modeling of machined depth in laser surface texturing of medical needles

The present paper addresses the experimental modeling of process parameters in laser surface texturing (LST) of medical needles. First, experiments were carried out based on Taguchi methodology. The laser process parameters considered during LST were the circumferential overlap, axial overlap and the overscan number. A second-order regression model of the machined depth for LST was developed based on the experimental results. Second, a predictive model for the machined depth based on least squares support vector machines (LS-SVM) with radial basis functions was constructed using the same experimental swatches. Grid search and leave-one-out cross-validation were used to determine the optimal parameters of the LS-SVM model. The comparison between the second-order regression model and the LS-SVM model was carried out. The experiments indicated that the LS-SVM model is capable of better predictions of the machined depth than the second-order regression model. The validity of the LS-SVM model has been checked through the creation of micro-channels with blended edges. It was found that the predicted profile was in a good agreement with the experimental profiles. The LS-SVM model can be used to predict machined geometry of the micro-channels on medical needles.     

Model for the prediction of 3D surface topography in 5-axis milling

The paper deals with the prediction of the 3D surface topography obtained in 5-axis milling in function of the machining conditions. For this purpose, a simulation model for the prediction of machined surface patterns is developed based on the well-known N-buffer method. As in sculptured surface machining the feed rates locally vary, the proposed model can be coupled to a feed-rate prediction model. Thanks to the simulation model of 3D surface topography, the influence of the machining strategy on resulting 3D surface patterns is analyzed through an experimental design. Results enhance the major influence of the tool inclination on 3D topography. Surface parameters used in the study are strongly affected by the variation of the yaw angle. The effect of the feed rate is also significant on amplitude parameters. Finally, the analysis brings out the interest of using surface parameters to characterize 3D surface topography obtained in 5-axis milling.     

基于ANFIS的铝合金铣削加工表面粗糙度预测模型研究

分析以往建立表面粗糙度预测模型方法的不足，采用自适应神经模糊推理系统(ANFIS)建立了铝合金铣削加工表面粗糙度预测模型。经检验，该模型预测精度高，泛化能力强，且可简便预测铣削参数对已加工表面的表面粗糙度的影响，有助于准确认识已加工表面质量随铣削参数的变化规律，为切削参数的优选和表面质量的控制提供了依据。     

Monitoring surface roughness of thin-walled components from steel C45 machining down and up milling

Article deals with problematic of milling thin wall components, than about study of surface roughness and analytical prediction of surface roughness Rz for variable geometrical parameters. First part is dedicated to research of realized experiments of manufacturing thin wall components, what was basis for designing of experiment. Experiment was conducted in two phases, where first was based on up milling and second on down milling for left and right side of thin wall components with thickness 10 mm. Subsequently, the surface roughness Rz was evaluated and was determinate mathematical equations for each type of milling, side and also for each depth. Final output of presented article is mathematical model of surface roughness Rz prediction for constant cutting conditions, but for variable geometrical parameters of thin wall components with thickness 10 mm.     

An efficient approach for milling dynamics modeling and analysis with varying time delay and cutter runout effect

An efficient approach for milling stability and surface location error (SLE) prediction with varying time delay and cutter runout effect is presented in this paper. Firstly, based on the tooth trochoid motion, the paper proposes a varying time delay model during cutter/workpiece engagement with taking cutter runout into account, establishes a milling dynamic model under arbitrary feed direction, and then derives the state transition matrix in one cutter rotation period by using the Cotes numerical integration formula. The milling stability of the dynamics system are obtained by using Floquet theory. According to the fixed point theory, the displacement response of the dynamic system and the method for solving the SLE are both developed. Later, a series of numerical and experimental works are conducted. The numerical verification shows that the proposed method can achieve a faster convergence rate and higher calculation efficiency than other previous methods. Meanwhile, the prediction of stability and SLE are in good agreement with the experimental results, and have a high accuracy for stability prediction when cutter runout and varying time delay considered. In the end, the numerical studies show that the milling stability and SLE strongly depends on the actual milling conditions, including milling parameters, cutter runout, cutter geometric parameters, and asymmetric structural dynamic parameters, which are helpful for milling process optimization.     

Surface roughness modeling in Laser-assisted End Milling of Inconel 718

Inconel 718 is a difficult-to-machine material while products of this material require good surface finish. Therefore, it is essential for the evaluation and prediction of surface roughness of machined Inconel 718 workpiece to be developed. An analytical model for the prediction of surface roughness under laser-assisted end milling of Inconel 718 is proposed based on kinematics of tool movement and elastic response of workpiece. The actual tool trajectory is first predicted with the consideration of overall tool movement, elastic deformation of tool, and the tool tip profile. The tool movements include the translation in feed direction and the rotation along its axis. The elastic deformation is calculated based on the previously established milling force prediction model. The tool tip profile is predicted based on the tool tip radius and angle. The machined surface profile is simulated based on the tool trajectory with elastic recovery, which is considered through the comparison between the minimum thickness and actual cutting thickness. Experiments are conducted in both conventional and laser-assisted milling under seven different sets of cutting parameters. Through the comparison between the analytical predictions and experimental measurements, the proposed model has high accuracy with the maximum error less than 27%, which is more accurate for lower feed rate with error less than 3%. The proposed analytical model is valuable for providing a fast, credible, and physics-based method for the prediction of surface roughness in milling process.