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。     

异形螺杆数控铣削刀具磨损智能建模研究

介绍了一种在线估算螺杆数控铣削中刀具磨损量的新方法。该方法基于螺杆铣削过程变切削参数的工况,提取了振动信号和功率信号的刀具磨损特征值,基于自适应神经—模糊推理系统建立了刀具磨损数学模型。实验证明,由此建立的刀具磨损模型能够排除切削参数变化的干扰,可以较好的反映加工中刀具磨损状态,同时也为具有时变切削参数特性的加工过程刀具磨损状态监控提供了新的研究方法。     

盘形成形铣刀的刀具磨损故障诊断方法研究

通过分析高速铣齿机的盘形成形铣刀及其铣削过程的特殊性、刀具磨损的信号采集和处理方法.选用电流法监测数控铣齿机的主轴电动机电流,采用BP神经网络方法对信号进行处理,最后实现对盘形成形铣刀的刀具磨损诊断.经实验证明,该铣刀的电流监测和基于BP神经网络数据处理的方法具有良好的诊断性能.     

采用深度学习的铣刀磨损状态预测模型

为提高刀具磨损监测的预测精度与泛化性能,研究了基于深度学习的铣刀磨损状态预测,提出了基于堆叠稀疏自动编码网络与卷积神经网络的两种预测模型。堆叠稀疏自动编码网络对特征向量进行降维并将其纳入分类器来实现预测,可避免特征选择对先验知识的依赖;卷积神经网络将铣削振动数据转化为小波尺度图并输入模型完成分类,精简了传统建模流程。最后将提出的两种模型与传统神经网络模型进行比较,验证了所提模型的效率与精度。     

异形螺杆铣削过程刀具磨损建模研究

介绍了一种螺杆铣削过程刀具磨损建模的方法。该方法针对螺杆加工中变切削参数的工况，提取了振动信号和功率信号的刀具磨损特征值，并建立了信号特征值与刀具磨损量之间的映射关系，从而得到刀具磨损模型。实验证明，由此建立的刀具磨损模型。能够排除切削参数变化的干扰，可以较好地反映加工中刀具磨损状态。同时也为具有时变切削参数特性的加工过程刀具磨损状态监控提供了新的研究方法。     

一种在线监测铣刀磨损量的新方法

提出了一种在线监测铣刀磨损量的新方法,该方法利用B样条神经网络建立不同刀具磨损状态下加工参数与切削力之间的映射关系。通过比较实时采集的切削力与不同刀具磨损值对应的切削力大小,可确定刀具的磨损状态,并利用建立的简化模型计算刀具的精确磨损值。试验结果表明,该方法消除了加工参数变化对特征的影响,简化了特征选取的方法,能够适应外部加工环境的变化,完全满足刀具状态监测系统的实用化需求。     

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。     

Continuous tool wear prediction based on Gaussian mixture regression model

The prediction of continuous tool wear process plays an important role in realizing adaptive control and optimizing manufacturing process so as to improve production efficiency and quality of the workpiece. However, the complexity of the tool wear process and the unpredictable disturbance during milling process make it difficult to realize robust and accurate estimation of the tool wear value. In this paper, the Gaussian mixture regression (GMR) model is proposed to realize continuous tool wear prediction based on features extracted from cutting force signal. The main characteristic of the GMR model is that the relationship between the tool wear value and the features is built by the combination of the Gaussian mixture model in which the variation of the training data is described by the probability density of the Gaussian distribution, and the wild data can be abandoned if its probability is small enough. To test the effectiveness of the proposed method, the experiment of titanium alloy milling was carried out, and the spectrum peak value corresponding to the harmonic of tooth passing frequency was extracted as the explanatory variables to predict the tool wear value. In addition, multiple linear regression, radius basis function, and back propagation neural network are also adopted to make a comparison with the GMR model. The analysis of four performance criteria shows that the GMR-based method is the most accurate among these methods.     

Prediction of tool wear using regression and ANN models in end-milling operation

Tool wear prediction plays an important role in industry for higher productivity and product quality. Flank wear of cutting tools is often selected as the tool life criterion as it determines the diametric accuracy of machining, its stability and reliability. This paper focuses on two different models, namely, regression mathematical and artificial neural network (ANN) models for predicting tool wear. In the present work, flank wear is taken as the response (output) variable measured during milling, while cutting speed, feed and depth of cut are taken as input parameters. The Design of Experiments (DOE) technique is developed for three factors at five levels to conduct experiments. Experiments have been conducted for measuring tool wear based on the DOE technique in a universal milling machine on AISI 1020 steel using a carbide cutter. The experimental values are used in Six Sigma software for finding the coefficients to develop the regression model. The experimentally measured values are also used to train the feed forward back propagation artificial neural network (ANN) for prediction of tool wear. Predicted values of response by both models, i.e. regression and ANN are compared with the experimental values. The predictive neural network model was found to be capable of better predictions of tool flank wear within the trained range.     

基于径向基函数神经网络的铣刀磨损监控技术的研究

提出一种基于径向基函数神经网络的铣刀磨损监控方法,径向基函数神经网络的输出是刀具磨损的具体值,这样有利于对刀具磨损进行各种实时补偿。实验表明,利用径向基函数神经网络进行状态识别可对小型立铣刀的磨损进行监控,能够取得良好的效果,同时证明RBF网络的训练速度优于BP网络。     

Detection process approach of tool wear in high speed milling

The cutting tool wear degrades the quality of the product in the manufacturing process, for this reason an on-line monitoring of the cutting tool wear level is very necessary to prevent any deterioration. Unfortunately there is no direct manner to measure the cutting tool wear on-line. Consequently we must adopt an indirect method where wear will be estimated from the measurement of one or more physical parameters appearing during the machining process such as the cutting force, the vibrations, or the acoustic emission, etc. The main objective of this work is to establish a relationship between the acquired signals variation and the tool wear in high speed milling process; so an experimental setup was carried out using a horizontal high speed milling machine. Thus, the cutting forces were measured by means of a dynamometer whereas; the tool wear was measured in an off-line manner using a binocular microscope. Furthermore, we analysed cutting force signatures during milling operation throughout the tool life. This analysis was based on both temporal and frequential signal processing techniques in order to extract the relevant indicators of cutting tool state. Our results have shown that the variation of the variance and the first harmonic amplitudes were linked to the flank wear evolution. These parameters show the best behavior of the tool wear state while providing relevant information of this later.     

The efficacy of back propagation neural network with delta bar delta learning in predicting the wear of carbide inserts in face milling

Face milling is a process predominantly affected by dynamic variation of cutting forces, thermo-mechanical shocks and vibration leading to catastrophic tool failure along with gradual wear of the inserts. Keeping in view the industrial importance of this process, it is necessary to devise suitable methods to predict in advance the onset of tool failure without grossly impairing the machining set-up and the job. Hence, the applicability of back propagation neural network with delta bar delta learning rule for faster convergence has been studied with the above objective. The multi sensor based tool condition monitoring strategy shows that the learning rate adaptation scheme combined with the selection of suitable process parameters drastically reduces the training time of the artificial neural network without dispensing with the prediction accuracy.     

基于主轴电流信号多特征融合的刀具磨损状态监测

为实现在正常生产条件下进行刀具磨损的长期在线监测,提出了基于主轴电流信号和粒子群优化支持向量机模型(PSO-SVM)的刀具磨损状态间接监测方法。首先对数控机床主轴电机电流信号进行分析,将与刀具磨损相关的主轴电流信号多个特征参数和EMD能量熵进行特征融合作为输入特征向量;其次,通过粒子群寻优算法(PSO)对支持向量机模型(SVM)参数进行优化,建立基于主轴电流信号融合特征和PSO-SVM理论的刀具磨损状态识别模型;最后,通过实验采集某立式加工中心主轴在刀具不同磨损状态下电流信号进行验证,并与传统SVM模型、BP神经网络模型进行了对比分析。结果表明,所提出的方法具有较高的准确率和较好的泛化能力。能够实现正常生产条件下对刀具磨损的长期在线监测。     

Milling tool wear diagnosis by feed motor current signal using an artificial neural network

In this paper, a Multi-layer perceptron (MLP) neural network was used to predict tool wear in face milling. For this purpose, a series of experiments was conducted using a milling machine on a CK45 work piece. Tool wear was measured by an optical microscope. To improve the accuracy and reliability of the monitoring system, tool wear state was classified into five groups, namely, no wear, slight wear, normal wear, severe wear and broken tool. Experiments were conducted with the aforementioned tool wear states, and different machining conditions and data were extracted. An increase in current amplitude was observed as the tool wear increased. Furthermore, effects of parameters such as tool wear, feed, and cut depth on motor current consumption were analyzed. Considering the complexity of the wear state classification, a multi-layer neural network was used. The root mean square of motor current, feed, cut depth, and tool rpm were chosen as the input and amount of flank wear as the output of MLP. Results showed good performance of the designed tool wear monitoring system.     

基于改进PCANet模型的铣刀磨损预测方法研究

铣刀健康状况直接影响实际生产加工过程,因此开展铣刀状态监测研究具有较大工程意义。以卷积神经网络为代表的深度学习模型已经逐渐用于监测加工过程中的刀具状态。但是这些模型的可解释性较差,预测结果的差异性也较大。作为一种新颖的卷积神经网络变种,主成分分析模型(Principal component analysis network, PCANet)的可解释性好,但是特征自监督学习能力有待提升,且相关应用案例较少。针对以上问题,拟对PCANet模型进行优化,进而提出了一种激活主成分分析-最大池化-支持向量回归(Activated PCANet with max pooling and support vector regression, APCANet-MP-SVR)模型,用于自适应提取敏感特征并准确预测刀具磨损情况。首先引入tanh激活函数,提高模型泛化能力;再采用最大池化层替代哈希编码和直方图用于特征选择,进一步降低冗余特征规模;最后建立支持向量回归模型实时预测刀具磨损值。应用案例充分证明了所提模型能够更好地用于加工现场刀具磨损值预测。     

TOOL CONDITION MONITORING USING ACOUSTIC EMISSION,SURFACE ROUGHNESS AND GROWING CELL STRUCTURES NEURAL NETWORK

The monitoring of tool wear is a most difficult task in the case of various metal-cutting processes. Artificial Neural Networks (ANN) has been used to estimate or classify certain wear parameters, using continuous acquisition of signals from multi-sensor systems. Most of the research has been concentrated on the use of supervised neural network types like multi-layer perceptron (MLP), using back-propagation algorithm and Radial Basis Function (RBF) network. In this article, a new constructive learning algorithm proposed by Fritzke, namely Growing Cell Structures (GCS) has been used for tool wear estimation in face milling operations, thereby monitoring the condition of the tool. GCS generates compact network architecture in less training time and performs well on new untrained data. The performance of this network has been compared with that of another constructive learning algorithm-based neural network, namely the Resource Allocation Network (RAN). For the sake of establishing the effectiveness of GCS, results obtained have been compared with those obtained using Multi Layer Perceptron (MLP), which is a standard and widely used neural network.     

基于神经网络多传感器融合的刀具摩损定量监测的研究

研究了前馈神经网络（ＦＮＮ）的自构造型学习算法，提出了基于神经网络多传感器融合的一般结构及刀具磨损监测方法，讨论了多传感器的选择、多传感器信号的采集与预处理以及多传感器信号的特征选择与正规化处理，并就铣削过程的刀具磨损监测进行了实验研究，结果表明，所提出的方法可获得９３％的识别率。     

FUZZY RADIAL BASIS FUNCTION (FRBF) NETWORK BASED TOOL CONDITION MONITORING SYSTEM USING VIBRATION SIGNALS

Thriving automation in industries leads to more research on the tool condition monitoring systems for better accuracy and fast recognition/evaluation of tool wear. Research on the applicability of the new advances in the soft-computing as well as in the signal processing fields is the inevitable consequence. In this work, a new soft-computing modeling technique, fuzzy radial basis function (FRBF) network has been applied to the prediction of drill wear using the vibration signal features. This work presents the wear prediction performance comparison of this new model with three other already tried and established soft-computing models, such as back propagation neural network (BPNN), radial basis function network (RBF) and normalized radial basis function network (NRBF), for both time-domain as well as wavelet packet approaches of feature extraction. Experimental results show that FRBF model with wavelet packet approach produces the best performance of predicting flank wear.     

基于主轴电流的铣削力间接监测方法

实时准确地监测铣削状态对于提高加工质量与加工效率具有重要意义,切削力作为重要的加工状态监测对象,因其监测设备昂贵且安装不便而受到限制,为此提出一种考虑刀具磨损的基于主轴电流的铣削力监测方法.首先基于切削微元理论建立了考虑后刀面磨损的铣削力模型,并通过铣削实验进行铣削力模型系数标定;然后对主轴电流与铣削力的关系进行理论建...     

A neural network based approach for background noise reduction in airborne acoustic emission of a machining process

Tool wear prediction has become an indispensable technique to prevent downtime in manufacturing and production processes. Airborne emission from a machining process using a low-cost microphone may provide a vital signal of tool health. However, the effect of background noise results in anomaly in data that may lead to wrong prediction of tool health. The paper presents an adaptive approach using neural networks for background noise filtration in acoustic signal for a turning process. Acoustic signal of a turning process is mixed with background noise from four different machines and introduced at different RPMs and feed-rate at a constant depth of cut. A comparison of Backpropagation neural network (BPNN), Self-organizing map and k-means clustering algorithm for noise filtration is investigated in this paper. In this regard, back-propagation neural network showed better performance with an average accuracy for all the four sources. It shows 100 % accuracy for grinding machine signal, 94.78 % accuracy for background signal from 3-axis milling machine, 45.57 % and 12.69 % for motor and 4-axis milling machine, respectively. Signal reconstruction is then done using Discrete cosine transform (DCT). The proposed technique shows a promising future for noise filtration in airborne acoustic data of a machining process.