基于BP神经网络和D-S证据理论的刀具磨损监测方法

将BP神经网络和D-S证据理论相结合的方法运用于刀具磨损监测中,采用小波包分解法对刀具磨损过程中产生的声发射信号进行特征提取,构建特征向量,利用BP神经网络识别判断刀具磨损状态;通过BP神经网络的输出结果和训练误差计算D-S证据理论的基本概率赋值,并用D-S证据理论对BP神经网络的识别结果进行决策级融合。实验结果表明:该方法避免了神经网络识别时的误诊,提高了整个刀具磨损监测系统识别的准确性和可靠性。     

Using neural network for tool condition monitoring based on wavelet decomposition

This paper presents a neural network application for on-line tool condition monitoring in a turning operation. A wavelet technique was used to decompose dynamic cutting force signal into different frequency bands in time domain. Two features were extracted from the decomposed signal for each frequency band. The two extracted features were mean values and variances of the local maxima of the absolute value of the composed signal. In addition, coherence coefficient in low frequency band was also selected as a signal feature. After scaling, these features were fed to a back-propagation neural network for the diagnostic purposes. The effect on tool condition monitoring due to the presence of chip breaking was studied. The different numbers of training samples were used to train the neural network and the results were discussed. The experimental results show that the features extracted by wavelet technique had a low sensitivity to changes of the cutting conditions and the neural network has high diagnosis success rate in a wide range of cutting conditions.     

Inspection of micro-tools at high rotational speeds

Micro-tools have been widely used in industry, primarily by biomedical and electronic equipment manufacturers. The life of these cutting tools is extremely unrpedictable and much shorter than conventional tools. Also, these miniature tools, with a diameter of less than 1 mm, cannot be inspected by an operator without the aid of magnifying glass.

In this paper, evaluation of the intensity variation of a reflected laser light beam from the cutting tool surfaces is proposed as a method of estimating cutting tool surface conditions. Various encoding methods, including wavelet transformations, were proposed to obtain a small and meaningful set of data from the intensity variation readings of one tool rotation. The encoded data are classified using a simple threshold method, Restricted Coulomb Energy (RCE), and Adaptive Resonance Theory (ART2)-type neural networks. The proposed encoding and classification approaches were tested with over one hundred sets of data.

The threshold method detects only severe tool damage. The RCE neural networks and graphical presentation of the encoded sets demonstrated the feasibility of the proposed monitoring technique and encoding methods. The ART2-type neural networks were found to be the best candidate for tool condition monitoring because of their self learning capability. Wavelet transformation-based encoding and ART2-type neural networks were found to be sensitive enough to recognize wear at the cutting edge.     

Machine ensemble approach for simultaneous detection of transient and gradual abnormalities in end milling using multisensor fusion

In a fully automated manufacturing environment, instant detection of the cutting tool condition is essential for the improved productivity and cost effectiveness. This paper studies a tool condition monitoring system (TCM) via machine learning (ML) and machine ensemble (ME) approach to investigate the effectiveness of multisensor fusion technique when machining 4340 steel with multilayer coated and multiflute carbide end mill cutter. In this study, 135 different features are extracted from multiple sensor signals of force, vibration, acoustic emission and spindle power in the time and frequency domain by using data acquisition and signal processing module. Then, a correlation-based feature selection technique (CFS) evaluates the significance of these features along with machining parameters collected from machining experiments. Next, an optimal feature subset is computed for various assorted combinations of sensors. Finally, machine ensemble methods based on majority voting and stacked generalization are studied for the selected features to classify not only flank wear but also breakage and chipping. It has been found in this paper that the stacked generalization ensemble can ensure the highest accuracy in tool condition monitoring. In addition, it has been shown that the support vector machine (SVM) outperforms other ML algorithms in most cases tested.     

A decision fusion algorithm for tool wear condition monitoring in drilling

Tool wear monitoring of cutting tools is important for the automation of modern manufacturing systems. In this paper, several innovative monitoring methods for on-line tool wear condition monitoring in drilling operations are presented. Drilling is one of the most widely used manufacturing operations and monitoring techniques using measurements of force signals (thrust and torque) and power signals (spindle and servo) are developed in this paper. Two methods using Hidden Markov models, as well as several other methods that directly use force and power data are used to establish the health of a drilling tool in order to avoid catastrophic failure of the drill. In order to increase the reliability of these methods, a decision fusion center algorithm (DFCA) is proposed which combines the outputs of the individual methods to make a global decision about the wear status of the drill. Experimental results demonstrate the effectiveness of the proposed monitoring methods and the DFCA.     

A brief review: acoustic emission method for tool wear monitoring during turning

Research during the past several years has established the effectiveness of acoustic emission (AE)-based sensing methodologies for machine condition analysis and process monitoring. AE has been proposed and evaluated for a variety of sensing tasks as well as for use as a technique for quantitative studies of manufacturing processes. This paper reviews briefly the research on AE sensing of tool wear condition in turning. The main contents included are:

1. The AE generation in metal cutting processes, AE signal classification, and AE signal correction.

2. AE signal processing with various methodologies, including time series analysis, FFT, wavelet transform, etc.

3. Estimation of tool wear condition, including pattern classification, GMDH methodology, fuzzy classifier, neural network, and sensor and data fusion.

A review of AE-based tool wear monitoring in turning is an important step for improving and developing new tool wear monitoring methodology.     

基于切削声和切削力参数融合的刀具磨损状态监测

针对单一传感器监测刀具磨损状态存在的不足,提出了将声传感方式和力传感方式综合利用,以人工神经网络作为多传感器信息融合的方法.在立式数控加工中心上铣削加工45<'#>钢调质试件,利用驻极体传声器和Kistler测力仪检测与刀具磨损相关的特征量,得出铣削声信号特征量LPCC的第6、7、8阶分量,X、Y向切削力以及绕z轴的力矩与刀具磨损密切相关.以这6个特征量作为神经网络的输入信号,利用有动量的梯度下降的BP算法建立了刀具磨损状态监测的多参数融合模型.研究结果表明神经网络输出值与实际测量值基本相符合,切削声和切削力特征融合后提高了识别刀具磨损程度的准确性和稳定性.     

数控镗刀磨损与破损的声发射监测法

通过分析声发射传感器采集的刀具磨损状态信号,提取出反映刀具磨损状态的特征向量MFCC系数及差分系数,然后利用隐马尔可夫模型进行信号处理,建立了检测镗刀刀具状态的监测系统。实验结果表明:在刀具的正常磨损阶段,该监测系统可以实现刀具大致磨损量的预报;在刀具破损或损坏情况下,能够及时监测和预报刀具损坏状态。这种监测方法可用于实时在线监测,为刀具的磨损监测提供了一条切实可行的途径。     

Cutting state monitoring in milling by a neural network

The application of a neural network to cutting state monitoring in face milling was introduced and evaluated on multiple sensor data such as cutting forces and vibrations. This monitoring system consists of a statistically based adaptive preprocessor (autoregressive (AR) time series modeling) for generating features from each sensor, followed by a highly parallel neural network for associating the preprocessor outputs (sensor fusion) with the appropriate decisions. AR model parameters were used as features, and the cutting states (normal, unstable and tool life end) were successfully detected by monitoring the evolution of model parameters during face milling. The proposed system offers fast operation through recursive preprocessing and highly parallel association, and a data-driven training scheme without explicit rules or a priori statistics. It appears proven on limited experimental data.     

网络多信道多源信息融合技术的改进研究

由于D-S融合算法采用的证据合成规则无法有效处理证据冲突问题,并且BP神经网络融合算法在样本波动情况下需要再次训练完成融合,容易出现局部最小值问题。因此,融合D-S融合算法和BP神经网络融合算法,提出基于上下文权值的多信道多源信息复合融合算法,其先基于检测数据的上下文,采用D-S融合算法和BP神经网络融合算法,对对应上下文内的数据进行融合处理,获取的融合结果被设置相应的权值,再将多个并行融合结果进行加权汇总,得到最终的融合结果,并同设置的阀值实施对比,获取最终的判决结果。实验结果说明,所提算法可有效处理证据冲突的融合问题,具有较高的准确性,融合效果佳。     

在线金属切削刀具磨损状态监测研究的回顾与展望Ⅲ:模式识别方法

刀具状态监测的目的是为了开发出实用的刀具状态监测设备,为了降低设备成本和提高监测的准确率,在监测信号的选择和特征提取的基础上,选择合适的模式识别方法至关重要。对近年来在学术期刊上公开发表的关于刀具磨损在线监测研究中所采用的主要模式识别方法作了简要的回顾与归纳,为后续研究者的快速入门及选择适当的模式识别方法提供参考。     

An approach based on current and sound signals for in-process tool wear monitoring

This paper presents a tool condition monitoring system (TCMS) for on-line tool wear monitoring in turning. The proposed TCMS was developed taking into account the necessary trade-off between cost and performance to be applicable in practice, in addition to a high success rate. The monitoring signals were the feed motor current and the sound signal. The former was used to estimate the feed cutting force using the least squares version of support vector machines (LS-SVM). Singular spectrum analysis (SSA) was used to extract information correlated with tool wear from the sound signal. The estimated feed cutting force and the SSA decomposition of the sound signal alone with the cutting conditions constitute the input data to the TCMS. Again LS-SVM was used to estimate tool condition and its reliability for on-line implementation was validated by experiments using AISI 1040 steel. The results showed that the proposed TCMS is fast and reliable for tool condition monitoring.     

基于小波神经网络的刀具状态监测系统的设计

利用能量分布特征提取方法和优化BP神经网络，提出一种基于小波变换和神经网络的刀具状态监测系统，利用该系统对四种刀具状态进行估计，结果与实际情况基本一致，证明该系统对于估计刀具状态是有效的。     

基于多模态集成卷积神经网络的数控机床齿轮箱故障诊断

针对数控机床齿轮箱在实际工作环境中负载多变且噪声干扰大、传统神经网络难以充分提取信号中的故障特征等问题,提出一种多模态集成卷积神经网络（MECNN）用于数控机床齿轮箱故障诊断。该方法将多模态融合技术与多个卷积神经网络结合,利用快速傅里叶变换方法将时域信号转换成频域信号;利用时域信号和频域信号对2个卷积神经网络进行训练,使模型能够分别从时域和频域2个角度提取特征,再将浅层特征融合;最后,将融合后的特征输入到卷积神经网络中进行故障特征的深度挖掘,并进行故障诊断。使用东南大学的齿轮箱数据集进行验证,设计了2种特征融合的方法并进行了对比。实验结果表明：在噪声下,MECNN模型用于故障诊断的准确性和鲁棒性均优于单一的时域CNN和频域CNN。     

Development of a tool failure detection system using multi-sensors

Tool monitoring and machine diagnosis in real machining have been crucial to the realization of fully automated machining. Also, the on-line detection technique of the tool breakage in machining should be supported. The effect of tool breakage is usually revealed from an abrupt change in the processed measurements, which is in excess of a threshold value. Although these techniques are generally effective for a specific cutting condition, they are often not sufficiently reliable for use in production due to the inability of single measurement to reflect tool breakage under various cutting conditions. In order to enhance the reliability of tool breakage signatures obtained from a single sensor, an integrated approach based on measurements from several sensors has been put forward. In this study, the tool breakage detection method using multi-sensors is proposed and the sensor fusion algorithm is developed to integrate and make decisions from data measured through the multisensors. Also, the performances of this scheme are compared and evaluated with real cutting process.     

基于BP神经网络的外圆磨削颤振在线识别和监测方法

为提高机床磨削加工过程中对颤振现象识别的能力,提出一种基于BP(back?propagation)神经网络模型的颤振识别方法。通过对加工过程中传感器采集到的高频声发射信号以及振动信号相关特征值的提取,获得关于颤振的多特征参数样本库,并用其对BP神经网络模型进行学习和训练,建立BP神经网络在线识别颤振的算法模型,实现对机床加工过程中是否发生颤振的在线监测和识别。试验结果表明:这种基于BP神经网络模型的颤振识别测试结果与磨削加工试验中的磨削颤振现象结果相符合。该方法能够有效地识别磨削加工过程中的颤振,并起到在线监测识别的作用。      

Evaluation of wear of turning carbide inserts using neural networks

Recent trends, being towards mostly unmanned automated machining systems and consistent system operations, need reliable on-line monitoring processes. A proper on-line cutting tool condition monitoring system is essential for deciding when to change the tool. Many methods have been attempted in this connection.Recently, artificial neural networks have been tried for this purpose because of its inherent simplicity and reasonably quick data-processing capability. The present work uses the back propagation algorithm for training the neural network of 5-3-1 structure. The technique shows close matching of estimation of average flank wear and directly measured wear value. Thus the system developed demonstrates the possibility of successful tool wear monitoring on-line.     

A multi-sensor strategy for tool failure detection in milling

A multi-sensor monitoring strategy for detecting tool failure during the milling process is presented. In this strategy, both cutting forces and acoustic emission signals are used to monitor the tool condition. A feature extracting algorithm is developed based on a first order auto-regressive (AR) model for the cutting force signals. This AR(1) model is obtained by using average tooth period and revolution difference methods. Acoustic emission (AE) monitoring indices are developed and used in determining the setting threshold level on-line. This approach was beneficial in minimizing false alarms due to tool runout, cutting transients and variations of cutting conditions. The proposed monitoring system has been verified experimentally by end milling Inconel 718 with whisker reinforced ceramic tools at spindle speeds up to 3000 rpm.     

Drilling wear detection and classification using vibration signals and artificial neural network

In automated flexible manufacturing systems the detection of tool wear during the cutting process is one of the most important considerations. This study presents a comparison between several architectures of the multi-layer feed-forward neural network with a back propagation training algorithm for tool condition monitoring (TCM) of twist drill wear. The algorithm utilizes vibration signature analysis as the main and only source of information from the machining process. The objective of the proposed study is to produce a TCM system that will lead to a more efficient and economical drilling tool usage. Five different drill wear conditions were artificially introduced to the neural network for prediction and classification. The experimental procedure for acquiring vibration data and extracting features in both the time and frequency domains to train and test the neural network models is detailed. It was found that the frequency domain features, such as the averaged harmonic wavelet coefficients and the maximum entropy spectrum peaks, are more efficient in training the neural network than the time domain statistical moments. The results demonstrate the effectiveness and robustness of using the vibration signals in a supervised neural network for drill wear detection and classification.     

Neural network solutions to the tool condition monitoring problem in metal cutting—A critical review of methods

A wide range of cutting tool monitoring techniques have been proposed and developed in the last decade, but only a few have found industrial applications, and a truly universally applicable system has still to be developed. In this paper a review of tool condition monitoring (TCM) systems, developed or implemented through application of neural networks, is provided. The review seeks to illustrate the extent of application of neural networks and the need for multiple source sensor signals in TCM systems. A critical analysis of methods is included and the trend in obtained results outlined.