基于神经网络的铣削大理石表面粗糙度预测模型

目的利用粒子群优化BP神经网络建立大理石加工表面粗糙度精确预测模型。方法首先采用不同切削参数进行铣削大理石试验,测量加工表面粗糙度值,同时对粒子群算法进行改进,使惯性权重按指数形式递减,并增加速度扰动系数,利用改进粒子群算法优化BP神经网络,建立铣削大理石表面粗糙度神经网络预测模型。其次使用部分试验数据来训练预测模型,使得到的网络参数让网络可以精确预测表面粗糙度。最后利用其余试验数据验证神经网络预测模型的准确性与可靠性。结果经过计算得到粒子群优化BP网络算法的预测模型归一化均方差为0.0501,最大相对误差为10.78%,且误差变化较为均匀。经验公式模型归一化均方差为0.1069,最大相对误差为39.64%,误差变化幅度较大。结论将神经网络模型与经验公式相比较,结果表明,所建网络模型具有较高的预测精度与较强的鲁棒性,对合理选择切削用量以得到理想表面粗糙度有一定参考价值。     

基于LS-DYNA仿真的射流加工参数分析

目的通过LS-DYNA对磨料射流冲蚀切削进行仿真,研究相关工艺参数对切削参数的影响。方法采用磨料水射流对Al_2O_3陶瓷进行了单点冲蚀仿真和切削仿真研究,其中水和磨料粒子采用SPH方法建模,氧化铝陶瓷工件采用FEM方法建模,并通过SPH-FEM耦合算法,实现射流冲蚀切削过程的仿真。结果分析射流冲蚀过程仿真和切削过程仿真可知,射流加工前期,由于射流中磨粒碰撞与反弹,使壁面成不规则"V"型。初始阶段,切深随计算时间呈线性增加,同时壁面对磨粒产生制约作用,从而使加工处的孔深基本不再增加。由于磨粒在冲蚀处壁面底部的冲蚀作用,使凹坑底部宽度增加并迅速趋于稳定。同时切削仿真与冲蚀仿真也存在一定区别,主要由于切削过程设定了移动速度。结论将仿真结果与实验结果进行比较可知,切削深度随着泵压的增大而成线性增大,切深随磨料流量的增大而增大,随靶距和横移速度的增大而减小。其中切深与磨料流量、靶距、横移速度均为非线性关系,工件最大切深与计算时间不呈线性关系增长。     

磨料水射流单颗磨粒侵彻的仿真研究

为了提高磨料水射流的材料去除率,采用ABAQUS软件建立磨料水射流单颗磨粒侵彻18CrNiMo7-6靶材的仿真模型,观察分析磨粒侵彻过程中靶材表面凹坑的变化情况;同时探究磨粒初始速度、磨粒直径、磨粒密度以及入射角度对凹坑形状及尺寸的影响规律,并根据材料去除体积优化磨料水射流工艺参数。结果表明：凹坑深度、材料去除体积随着磨粒初始速度、磨粒直径、磨粒密度、入射角度的增加而增大,凹坑宽度随着磨粒初始速度、磨粒密度的增加先增大而后保持在最大宽度,随着磨粒直径的增加而增大,随着入射角度的增加而减小。以最大材料去除体积为目标,此仿真方案中最优参数为：磨粒初始速度为400 m/s,磨粒密度为12 000 kg/m3,入射角度90°,磨粒直径为1 mm。     

基于GPU并行计算的风沙流SPH数值分析

在研究风沙流动方面,光滑粒子流体动力学方法(SPH)的无网格性有着独有的优势。利用SPH方法研究风沙流动时,需要将整个计算区域离散成数量庞大的单个粒子,因此计算规模大、计算效率低。为提高SPH方法的计算效率,采用支持并行计算的CUDA平台,利用GPU大规模并行计算技术,实现SPH方法数值模拟时的加速运算。以二维气沙两相耦合模型作为数值算例,利用GPU并行计算详细分析颗粒群的运动规律。比较在不同粒子数下CPU与GPU的计算效率以及GPU线程数对计算效率的影响。对所得结果进行统计分析后,得到了单颗沙粒的典型抛物线形和变异的跃移轨迹。模拟结果证明:SPH-GPU并行计算技术能够应用在风沙流结构的数值模拟研究中。     

Tool flank wear prediction in CNC turning of 7075 AL alloy SiC composite

Flank wear occurs on the relief face of the tool and the life of a tool used in a machining process depends upon the amount of flank wear; so predicting of flank wear is an important requirement for higher productivity and product quality. In the present work, the effects of feed, depth of cut and cutting speed on flank wear of tungsten carbide and polycrystalline diamond (PCD) inserts in CNC turning of 7075 AL alloy with 10 wt% SiC composite are studied; also artificial neural network (ANN) and co-active neuro fuzzy inference system (CANFIS) are used to predict the flank wear of tungsten carbide and PCD inserts. The feed, depth of cut and cutting speed are selected as the input variables and artificial neural network and co-active neuro fuzzy inference system model are designed with two output variables. The comparison between the results of the presented models shows that the artificial neural network with the average relative prediction error of 1.03% for flank wear values of tungsten carbide inserts and 1.7% for flank wear values of PCD inserts is more accurate and can be utilized effectively for the prediction of flank wear in CNC turning of 7075 AL alloy SiC composite. It is also found that the tungsten carbide insert flank wear can be predicted with less error than PCD flank wear insert using ANN. With Regard to the effect of the cutting parameters on the flank wear, it is found that the increase of the feed, depth of cut and cutting speed increases the flank wear. Also the feed and depth of cut are the most effective parameters on the flank wear and the cutting speed has lesser effect.     

基于BP神经网络与遗传算法的固结磨具制作工艺参数优化

目的 利用BP神经网络技术与遗传算法寻找固结磨具制作最优工艺参数组合,实现固结磨具制作工艺参数的快速寻优.方法 设计磨粒粒径、磨粒质量分数、成型压力、烧结温度的正交工艺参数表,按正交表工艺参数制作蓝宝石晶片加工用的Cr2O3固结磨具,并且设计不同固化温度下制作的固结磨具的硬度与抗压强度测试试验,验证自制的固结磨具加工的...     

Process geometry modeling with cutter runout for milling of curved surfaces

Prediction of cutting forces and machined surface error in peripheral milling of curved geometries is non-trivial due to varying workpiece curvature along tool path. The complexity in this case, arises due to continuously changing process geometry as workpiece curvature varies along tool path. In the presence of cutter runout, the situation is further complicated owing to changing radii of cutting points. The present work attempts to model process geometry in machining of curved geometries and in the presence of cutter runout. A mathematical model computing process geometry parameters which include cutter/workpiece engagements and instantaneous uncut chip thickness in the presence of cutter runout is presented. The developed model is more realistic as it accounts for interaction of cutting tooth trajectory with that of preceding teeth trajectories in computing process geometry. Computer simulation studies carried for this purpose has shown that it is essential to account for teeth trajectory interactions for accurate prediction of process geometry parameters. This aspect is further confirmed with machining experiments, which were conducted to validate this aspect. From the outcomes of present work, it is clearly seen that the computation of process geometry during machining of curved geometries and in presence of cutter runout is not straightforward and requires a systematic approach as presented in this paper.     

铣削7075铝合金表面粗糙度试验研究

为进一步探究加工参数与7075铝合金表面粗糙度之间的变化关系。开展铣削7075铝合金表面粗糙度试验,基于单因素试验结果分析加工参数与表面粗糙度之间的影响规律,基于含有交互作用的正交试验结果,分析各加工因素最优参数水平,构建表面粗糙度二、三阶响应曲面预测模型。研究表明:表面粗糙度随着切削速度、进给量、切削深度的逐渐增加而增大;表面粗糙度各因素的最优参数水平为A2B1C1;对比分析F值、复相关系数,表面粗糙度三阶响应曲面预测模型优于二阶。确定的最优预测模型为深入研究加工参数与表面粗糙度之间变化关系奠定了理论基础。     

自蔓延高温合成多孔NiTi合金孔隙的SVR预测

根据自蔓延高温合成法(SHS)制备多孔NiTi合金孔隙试验所获得的实测数据集,应用基于粒子群算法(PSO)寻优的支持向量回归(SVR)方法,建立不同反应参数(温度,粒度和压坯密度)下合成的多孔NiTi合金孔隙的SVR预测模型,并与基于误差反向传播神经网络(BPNN)回归模型的预测结果进行比较。结果表明：在相同的训练与测试样本集下所获的SVR预测结果的平均绝对百分误差(MAPE)比BPNN预测模型的要小,其预测精度更高,预测效果更好;SVR-LOOCV预测的MAPE也比BPNN略小,且其预测结果的相关系数达到了0.999。因此,该方法是一种预测SHS法制备多孔NiTi合金孔隙的有效方法,可为SHS合成多孔NiTi提供理论指导     

Nano-precision measurement of diamond tool edge radius for wafer fabrication

In this paper, a non-destructive nano-precision measurement method for diamond tool cutting edge radius is presented. The basis of the method is that the profile of a tool cutting edge can be copied by indenting the tool cutting edge into the surface of a selected material, and that the copy of the profile can be measured at nano-precision level using AFM. The selected material elastic error compensation coefficient has to be determined to cancel out the effect of elastic spring-back. Copper was selected as the indentation piece material due to its (1) high rigidity and high density, (2) large Young’s modulus and (3) low yield strength. The elastic error compensation coefficient for the copper material is determined through the indentation of a tungsten carbide tool edge on the copper surface. By comparing the actual tool edge radius measured using scanning electron microscope (SEM) on the sectional view of the tungsten carbide tool with the one measured from the copied profile of the tool edge on the copper surface, the coefficient is obtained. Three diamond tool edge radii were obtained using the proposed method. Analysis is given for the accuracy of the proposed method, showing that as far as the error elastic compensation coefficient is consistent with the copper material used, the only source of errors with the measurement will come from the device for measuring the indented profile on the surface.     

基于SPH法的钢轨打磨单颗磨粒磨削仿真

为研究钢轨打磨过程中材料的去除机理,采用光滑粒子流体动力学（SPH）的方法,仿真模拟钢轨打磨过程中单颗磨粒的切削过程,分析单颗磨粒几何形状、切削深度、负前角对打磨磨削过程中切削力、切削力比的变化规律及工件材料应力、变形情况的影响。结果表明:由于单颗磨粒的推挤作用,工件材料流动后形成毛刺和磨屑,而棱锥形磨粒可以获得较好的磨削加工表面;切削力随磨粒切削深度的增加而增大;磨粒负前角增大时,切削力和切削力比都随之增大,且负前角越大磨屑呈越明显的锯齿状。      

Milling error prediction and compensation in machining of low-rigidity parts

The paper reports on a new integrated methodology for modelling and prediction of surface errors caused by deflection during machining of low-rigidity components. The proposed approach is based on identifying and modelling key processing characteristics that influence part deflection, predicting the workpiece deflection through an adaptive flexible theoretical force-FEA deflection model and providing an input for downstream decision making on error compensation. A new analytical flexible force model suitable for static machining error prediction of low-rigidity components is proposed. The model is based on an extended perfect plastic layer model integrated with a FE model for prediction of part deflection. At each computational step, the flexible force is calculated by taking into account the changes of the immersion angles of the engaged teeth. The material removal process at any infinitesimal segment of the milling cutter teeth is considered as oblique cutting, for which the cutting force is calculated using an orthogonal–oblique transformation. This study aims to increase the understanding of the causes of poor geometric accuracy by considering the impact of the machining forces on the deflection of thin-wall structures. The reported work is a part of an ongoing research for developing an adaptive machining planning environment for surface error modelling and prediction and selection of process and tool path parameters for rapid machining of complex low-rigidity high-accuracy parts.     

Robust and accurate prediction of thermal error of machining centers under operations with cutting fluid supply

A novel temperature measuring system named LATSIS was proposed to realize a robust and accurate prediction of the thermal deformation of machining centers, even under external disturbances such as cutting fluid supply. LATSIS enables a drastic increase in the number of sensors employed for measuring the temperature of the machine tool. Thus, the entire temperature distribution can be obtained by interpolating the measured temperature 3-dimensionally without calculating the heat conduction. A set of experiments was conducted in which the LATSIS was employed to predict the TCP error. A total of 284 sensors were placed on the machining center, and the TCP error was predicted based on the measured temperature for the situation with/without the cutting fluid supply. The results of the prediction showed good agreement with the measured TCP error even during the initial transient temperature change as well as in the cooling phase after the machine halt. The TCP error with the cutting fluid supply is accurately predicted. LATSIS was proven to be a robust and accurate method for predicting the thermal deformation of machine tools, and is a promising technology for future manufacturing systems.     

基于PSO-ELM的316L不锈钢细长管磁粒研磨内表面粗糙度预测模型

针对316L不锈钢细长管磁粒研磨加工过程中,最佳工艺参数难以选择,以及加工后对工件内表面粗糙度（Ra）的预测问题,将影响磁粒研磨316L不锈钢细长管内表面粗糙度的四个工艺参数作为输入值,内表面粗糙度作为输出值,构建粒子群（PSO）优化极限学习机（ELM）模型来预测316L不锈钢细长管内表面粗糙度,利用PSO对工艺参数进行全局寻优,获得最佳工艺参数组合,最后通过试验与预测结果进行对比。构建的PSO-ELM表面粗糙度预测模型拟合优度R2为0.984 8,绝对误差（MAE）为0.013 4,均方根误差（RMSE）为0.021 4。得到的最佳工艺参数组合为：主轴转速2 389.011r/min,进给速度3.167 mm/s,磨料粒径216.185μm,加工时间35.856 min,预测Ra为0.178μm。对工艺参数进行调整,试验得到的Ra为0.182μm,与预测值相比误差为2.24%。基于PSO-ELM方法构建316L不锈钢细长管内表面粗糙度预测模型,实现对工件内表面粗糙度的精确预测,应用粒子群方法得到最佳工艺参数组合,提高了磁粒研磨316L不锈钢细长管的加工效率。     

基于MATLAB神经网络的切削力预测

借助MATLAB人工神经网络,对切削力预测进行了研究。通过比较快速BP算法和LM算法在网络训练时的收敛速度,确定了网络的结构和工具函数,并分析了影响切削力预测精度的因素,实现了切削力的精确预测。其研究结果为车削零件加工质量的物理仿真以及加工参数的优化选择提供了依据。     

Prediction of cutting forces in milling of circular corner profiles

This paper proposes an approach to predict the cutting forces in peripheral milling of circular corner profiles in which varying radial depth of cut is encountered. The geometric relationship between an end mill and the corner profile is investigated and a mathematical model is presented to describe the different phases of the cutter/workpiece contact. The milling process for circular corner is discretized into a series of steady-state cutting processes, each with different radial depth of cut determined by the instantaneous position of the end mill relative to the workpiece. A time domain analytical model of cutting forces for the steady-state machining conditions is introduced to each segmented process for the cutting force prediction. The predicted cutting forces can be calculated in terms of tool/workpiece geometry, cutting parameters and workpirece material property, as well as the relative position of the tool to workpiece. Experiments are conducted and the measured forces are compared to the predictions for the verification of the proposed method.     

基于GA-BP的6061Al切削参数优化

针对6061Al铣削中表面粗糙度预测精度低、切削参数选择不合理的问题,提出一种基于遗传神经网络与遗传算法结合的优化模型,对6061Al切削参数进行优化。采用遗传神经网络(GA-BP)构建表面粗糙度预测模型;基于表面粗糙度预测,以材料去除率为目标函数构建切削参数优化模型;利用遗传算法进行优化求解,对6061Al切削参数进行优化。研究结果表明:所建预测模型表面粗糙度预测精度在97%以上;同时,优化模型能优化6061Al切削参数,达到较好的全局寻优效果,为铝合金工件铣削加工切削参数优化提供参考。     

SPH-FEM simulations of microwave-treated basalt strength

Microwave precondition has been highlighted as a promising technology for softening the rock mass prior to rock breakage by machine to reduce drill bit/cutter wear as well as inverse production rate. To numerically explore the effect of numerical parameters on rock static strength simulation, and determine the numerical mechanical parameters of microwave-treated basalts for future drilling and cutting simulations, numerical models of uniaxial compression strength (UCS) and Brazilian tensile strength (BTS) were established with the coupling of smoothed particle hydrodynamics and finite element method (SPH-FEM). To eliminate the large rock strength errors caused by microwave-induced damage, the cohesion and internal friction angle of microwave-treated basalt specimens with the same microwave treatment parameters were calibrated based on a linear Mohr-Coulomb theory. Based on parametric sensitivity analysis of SPH simulation of UCS and BTS, experimental UCS and BTS values were simultaneously captured according to the same set of calibrated cohesion and internal friction angle data, and the UCS modeling results are in good agreement with experimental tests. Furthermore, the effect of microwave irradiation parameter on the basalt mechanical behaviors was evaluated.     

基于核偏最小二乘法的动态预测模型在铜转炉吹炼中的应用

为实现铜转炉吹炼过程中的关键操作参数的准确预测,构造一种基于核偏最小二乘法的动态预测模型,并提出一种适用于动态建模的在线式异常样本剔除方法。该动态预测模型使用滑动窗方法不断更新建模数据,再利用核偏最小二乘法对动态模型的参数进行辨识,最后根据反馈的前次计算误差对本次预测值进行修正。仿真研究结果表明：该动态预估模型具有较好的泛化能力和较强的鲁棒性,并具有较好预测精度（风量预测的相对均方根误差小于10%,氧量预测的相对均方根误差小于19%）。目前,该预测模型被用于某转炉的吹炼辅助决策系统中。     

热喷涂Cr/Al2O3复合粉末的制备及性能研究

目的获得热喷涂用包覆型Cr/Al_2O_3复合粉末的制备工艺,探究工艺参数对热喷涂粉末结构及性能的影响规律。方法将纳米Al_2O_3水分散液与粘结剂混合润湿形成胶状液体,然后使其在核心粒子Cr表面团聚直接得到陶瓷相包覆金属相的复合颗粒,确定最佳制备工艺参数,并通过扫描电子显微镜(SEM)、能谱仪(EDS)、X射线衍射(XRD)和霍尔流速与松装密度计研究工艺参数对复合粉末结构和性能的影响。结果在核-壳结构复合粉末制备过程中,加入一定量的粘结剂和减少包覆次数可以改善包覆效果,最终制得的包覆型颗粒壳层厚度可以达到25μm。随着Cr含量的增大,包覆效果有所下降,但粉末流动性变好,松装密度值提高。初始Cr粒度增大,包覆效果增强,颗粒球形度改善,但流动性和松装密度变化不大。结论机械包覆Cr/Al_2O_3复合粉末的最佳制备工艺参数为加入质量分数为5%的粘结剂进行一次包覆,该方法制得的复合粉末粒度分布均匀,流动性和松装密度值良好,适合热喷涂。