基于BP神经网络的机器人砂带磨削表面粗糙度研究

为提高机器人砂带磨削工件表面粗糙度的预测精度,采用基于BP神经网络方法进行研究,进行机器人砂带磨削铝合金板材试验,基于试验结果采用BP神经网络建立各工艺参数与工件表面粗糙度之间的预测模型。对该模型进行仿真预测,并通过试验验证该模型的预测精度。结果表明该模型预测精度高,可以预测不同工艺参数磨削后的工件表面粗糙度,实现了机器人砂带磨削铝合金板材工艺参数的优化。     

数控铣削加工工艺参数优化方法综述

介绍了铣削力、铣削用量等数控铣削加工工艺参数,分析了材料去除率、表面粗糙度、能耗、铣刀颤振等工艺指标,并给出了数控铣削加工工艺参数的优化目标、优化方法、现有试验研究,以及近似模型。所做研究可以为数控铣削加工工艺参数的选择和优化提供理论参考。     

GA-BP神经网络预测钛合金表面粗糙度

为了有效预测铣削加工中钛合金工件的表面粗糙度,建立了以切削速度、进给量、径向切深、轴向切深为输入参数,表面粗糙度为输出参数的预测模型。该预测模型将遗传算法与BP神经网络结合起来,使用遗传算法优化BP神经网络的初始权值和阈值,进行铣削实验获得实验数据,并对神经网络进行训练,最终获得预测模型。通过对比分析GA-BP预测模型、BP预测模型、线性回归预测模型的预测精度,得出GA-BP预测模型具有相对较好的预测精度,证明该预测模型是有效的。     

数控加工铣削试验数据的清洗与处理

样本数据作为神经网络模型训练的导师,其质量直接影响了神经网络的学习能力以及神经网络模型的预测能力。以DMC60H为试验平台,以壳体类铝合金零件加工为研究对象,提取数控铣削加工试验数据;通过对数控铣削参数试验数据的分析与研究,提出了试验数据与样本数据的处理原则,分析了验证数据的构成以及所占数量。实现了样本数据的优化,并同时剔除了样本数据中的错误信息。     

高速铣削工件表面粗糙度的预测

表面粗糙度是衡量工件表面质量的重要指标。采用正交试验方法,利用圆环面铣刀对模具钢NAK80进行了高速铣削试验,测量了不同工艺参数下的工件表面粗糙度。将试验结果与人工智能中的BP神经网络结合,建立了表面粗糙度预测模型,用于预测不同主轴转速、进给速度、切削深度、切削行距、刀具倾角时被加工工件的表面粗糙度,并通过MATLAB图形用户界面设计了表面粗糙度预测软件。结果表明,该预测模型及其封装后的软件可用于加工前工件表面粗糙度的预测。     

数字孪生驱动的数控铣削刀具磨损在线监测方法

针对数控铣床不断老化导致刀具磨损预测模型误差较大,加工过程中动态数据难以在线采集等问题,提出一种数字孪生驱动的刀具磨损在线监测方法。采用神经网络对加工过程中的多源数据进行特征提取,建立考虑机床老化的刀具磨损时变偏差量化模型,并在此基础上提出数控铣削刀具磨损的在线预测方法;开发了面向刀具磨损的数控铣削数字孪生系统,在线感知加工过程中的动态数据并实时仿真刀具磨损过程;最后,将该方法应用于实际加工中并与其他的预测方法进行了对比,结果表明该方法有效降低了机床老化带来的误差,实现了刀具磨损的精确预测。     

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BP神经网络的车削Ti6Al4V高温钛合金表面粗糙度预测

针对Ti6 Al4 V高温钛合金轴类零件的车削加工过程,设计了Ti6Al4V高温钛合金全因素车削实验,对不同车削参数下表面粗糙度的数值进行了测量,然后以车削参数为输入变量,以表面粗糙度为响应,建立了神经网络预测模型。然后在同等条件下实施验证实验,验证了神经网络模型的预测精度。结果显示,文中研究建立的神经网络模型可以根据切削变量准确预测机械零件表面粗糙度。研究结果可为现场生产选择合适的切削参数提供理论指导,从而实现机械零件表面粗糙度的预测和控制。     

基于过程集成优化的复杂曲面数控铣削性能分析

针对复杂自由曲面变曲率、大扭曲的特征及其铣削性能难预测的问题,提出加工过程的集成优化的切削性能分析方法,在曲面多轴铣削工作流程中综合评价和提高切削加工效率和质量。建立了自由曲面体零件多轴加工集成优化铣削模型,集成切削加工刀位轨迹计算、切削仿真与机床运动仿真、切削力预测、工艺参数优化工作流程及其输入输出文件,实时从输出文件中解析提取计算结果参数,有效分析预测切削参数与切削力对加工效率和质量的影响,实现复杂自由曲面铣削过程的集成与全局优化。将该方法应用于大型混流式水轮机叶片的数控铣削性能分析,并与生产数据进行对比,进一步验证了所提加工过程集成优化方法可有效分析和预测大型自由曲面的数控铣削性能。     

基于神经网络的电化学加工表面粗糙度预测与加工参数正交优化

电化学加工的表面粗糙度与加工电流、加工间隙、电解液温度、加工时间、电解液配比等工艺参数密切相关,而这些工艺参数与工件表面粗糙度之间为复杂的非线性关系,建立其关联一直是电化学加工中的难题。以BP神经网络为基本工具,建立了加工参数与表面粗糙度之间关系的数学模型,利用实验数据训练网络,结果表明可实现较小的预测误差;应用正交法分析实验数据,实现了可使表面粗糙度参数变化幅度较大的加工参数的优化配置。     

Roughness modeling and optimization in CNC end milling using response surface method: effect of workpiece material variation

Influence of machining parameters, viz., spindle speed, depth of cut and feed rate, on the quality of surface produced in CNC end milling is investigated. In the present study, experiments are conducted for three different workpiece materials to see the effect of workpiece material variation in this respect. Five roughness parameters, viz., centre line average roughness, root mean square roughness, skewness, kurtosis and mean line peak spacing have been considered. The second-order mathematical models, in terms of the machining parameters, have been developed for each of these five roughness parameters prediction using response surface method on the basis of experimental results. The roughness models as well as the significance of the machining parameters have been validated with analysis of variance. It is found that the response surface models for different roughness parameters are specific to workpiece materials. An attempt has also been made to obtain optimum cutting conditions with respect to each of the five roughness parameters considered in the present study with the help of response optimization technique.     

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.     

基于多源同步信号与深度学习的刀具磨损在线识别方法

为提高刀具状态监测系统的实用性、避免实际加工过程中工序变换产生的信号干扰,提出一种基于多源同步信号与深度学习的刀具磨损在线识别方法。该方法利用自动触发的方式实现了机床运行在特定工序时的刀具振动、主轴功率、数控系统参数等多源信号的同步在线采集,保证信号同步性的同时有效避免了因工序变换而产生的信号波动干扰;进一步利用高频振动特征实现了 “切削过程”与“切削间隙”采集样本的准确划分,并基于皮尔逊积矩相关系数筛选出强关联特征,保证了多源监测信号融合样本的可用性;最后基于一维卷积神经网络建立了刀具磨损在线识别模型。实验结果表明,该方法无论从识别精度还是诊断效率,均能实现实际加工过程中刀具磨损状态的在线识别。     

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.     

模具钢铣削中刀具磨损的试验研究

以面铣刀刀片磨损为研究对象,结合类神经网络系统建构高速数控铣削加工的预测模型。以加工参数为模型输入条件,刀腹磨耗为输出条件。采用多因素试验方法,选择切削速度、进给速度、切削深度三个试验参数,利用直交表式的试验计划法设计试验点。依照试验点铣削工件后再测量刀具加工后的刀腹磨耗量,进而求得倒传递网络所需的36组训练范例与11组验证数据。刀腹磨耗预测模式是利用类神经网络中的倒传递网络原理,以田口法求得倒传递网络参数的最优值。试验结果显示,刀腹磨耗随着切削速度、进给速度、切削深度增加而上升。铣削模具钢后,刀具磨耗预测值的平均误差为4.72%,最大误差为11.43%,最小误差为0.31%。整体而言,类神经网络对于铣削加工可进行有效预测。     

基于多误差源耦合的五轴数控铣床加工误差综合预测及评判

在综合考虑机床动静态多种误差源的基础上,建立了各运动轴伺服运动模型和多体联动模型,给出了刀具的实际运动位置和姿态,基于包络理论求解了曲面加工实际成形面,对比理想数学模型,对加工误差进行了综合预测和评判。以复杂非可展曲面--S试件为例,给出了S试件的铣削精度构建方法,分析了机床动态因素(位置环、速度环等)对零件铣削精度的影响,并通过切削实验后的数据回归分析予以验证。建立了基于神经网络的机床铣削误差辨识模型,用于评估机床加工后的状态。该平台的搭建为实现大型、关键零件的加工精度预测和保障提供了技术支撑。      

An investigation into material-induced surface roughness in ultra-precision milling

This paper presents a theoretical and experimental investigation into the effect of the workpiece material on surface roughness in the ultra-precision milling process. The influences of material swelling and tool-tip vibration on surface generation in ultra-precision raster milling are studied. A new method is proposed to characterize material-induced surface roughness on the raster-milled surface. A new parameter is defined to characterize the extent of surface roughness profile distortion induced by the materials being cut. An experiment is conducted to compare the proposed method with surface roughness parameters and power spectrum density analysis method by machining three different workpiece materials. The results show that the presence of elastic recovery improves the surface finish in ultra-precision raster milling and that, among the three materials being cut in the experiment, aluminum bronze has the greatest influence on surface finish due to its highest elastic recovery rate and hardness. The results also show that, in the case of faster feed rates, the proposed method more efficiently characterizes material-induced surface roughness.     

Investigation of the effect of machining parameters on the surface quality of machined brass (60/40) in CNC end milling—ANFIS modeling

Brass and brass alloys are widely employed industrial materials because of their excellent characteristics such as high corrosion resistance, non-magnetism, and good machinability. Surface quality plays a very important role in the performance of milled products, as good surface quality can significantly improve fatigue strength, corrosion resistance, or creep life. Surface roughness (Ra) is one of the most important factors for evaluating surface quality during the finishing process. The quality of surface affects the functional characteristics of the workpiece, including fatigue, corrosion, fracture resistance, and surface friction. Furthermore, surface roughness is among the most critical constraints in cutting parameter selection in manufacturing process planning. In this paper, the adaptive neuro-fuzzy inference system (ANFIS) was used to predict the surface roughness in computer numerical control (CNC) end milling. Spindle speed, feed rate, and depth of cut were the predictor variables. Experimental validation runs were conducted to validate the ANFIS model. The predicted surface roughness was compared with measured data, and the maximum prediction error for surface roughness was 6.25 %, while the average prediction error was 2.75 %.     

A dynamic surface roughness model for face milling

This paper presents a newly developed mathematical model for surface roughness prediction in a face-milling operation. The model considers the static and the dynamic components of the cutting process. The former includes of cutting conditions as well as the edge profile and the amount of runout of each insert set into a cutter body. The latter introduces the dynamic characteristics of the milling process. It is verified that such a model predicts the maximum or the arithmetic mean surface roughness value through the cutting experiments. The model can evaluate the surface texture of the precision parts machined with face milling.