基于铣削力参数化模型的铣刀非均匀磨损状态监测方法研究

在航空结构件钛合金零件数控加工中,刀具非均匀磨损状态对工件的最终质量影响很大。为了及时发现并控制因刀具非均匀磨损导致的异常加工状态,对钛合金加工刀具非均匀磨损状态监测方法进行了研究。建立了基于刀具刃线参数化模型的铣削力参数化模型,实现了对钛合金加工刀具非均匀磨损状态的监测,解决了零件单件或首件加工中样本数据缺失条件下的钛合金加工刀具非均匀磨损状态准确监测难题。     

HMM技术在数控镗刀状态监测中的应用

通过对声发射传感器采集的刀具磨损状态信号进行分析,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HMM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,该监测系统在刀具的正常磨损阶段,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。这种监测方法能够进行实时在线监测,为刀具的磨损监测提供了一条切实可行的途径。     

基于粒子群优化的螺纹刀具磨损状态监测模型

螺纹刀具的状态监测是制造加工中非常重要的问题.由于刀具振动信号具有复杂非线性、强耦合等关系,常用的基于支持向量机(SVM)的刀具监测模型由于参数的设置极其依赖人为经验,设置不当会导致监测的识别率不高,在刀具磨损状态判别中收到了限制.针对此难题,依据螺纹刀具的振动特性,结合改进的粒子群算法(PSO),采用异步更新学习因子策略实现刀具状态监测模型优化.结果表明,优化后的PSO-SVM刀具状态监测模型能够有效对SVM的关键参数进行寻优,异步更新学习因子也可加强模型在迭代后期的寻优能力,从而提高刀具状态监测识别的精度.     

精密数控机床铣刀故障监测方法的研究

刀具在加工过程中不可避免的存在着磨损和破损现象,刀具的消耗直接导致工件精度下降和生产成本增加。开展了一系列实验,深入研究刀具状态监测方法,构建了新型铣削过程刀具磨损监测试验系统。通过振动传感器和声发射传感器对铣削过程中不同磨损程度刀具的信号进行检测、采集、分析。选择对刀具磨损状态反映敏感的特征量。采用BP神经网络,建立刀具磨损特征向量与刀具磨损状态之间的非线性映射关系。     

基于小波分析刀具破损检测

刀具切削状态的在线监测是自动化加工技术中的难题,国内外已有许多专家和学者进行了深入研究。针对目前刀具切削状态在线监测中出现的问题,提出了一种利用小波分析技术提取刀具状态信号特征量,对刀具的破损进行预报的监测系统。经大量试验和使用,验证了其可行性和实用性。     

刀具切削状态的电机电流监测新方法

本文从实时在线监测刀具切削状态的应用出发,研制了一种新型的电机电流拾取方法和传感装置,并分析了刀具在加工过程中切削力,电机电流与刀具切削状态的关系,同时,研究了利用电机电流信号进行刀具切削状态的监测原理,实验证明了其方法的可靠性和实用性。     

基于小波包和BP神经网络的刀具磨损状态识别

刀具磨损状态影响金属切削过程,因此监测刀具磨损状态对提高产品质量有着重要的意义。设计刀具磨损状态监测系统,利用传感器采集刀具振动信号,通过小波包对振动信号进行数据分析,并把不同频段的能量值作为刀具磨损状态的特征值,建立BP神经网络,从而在刀具磨损状态和振动信号特征向量之间建立映射关系,完成刀具磨损状态的监测。利用C++Builder和Matlab软件混合编程实现了系统的功能。试验表明,系统运行良好,能够对刀具磨损状态进行正确识别。     

基于表面二维PCA重构图像的刀具磨损分形特征研究

根据加工表面纹理图像与刀具几何形状之间的内在联系,提出利用计算机视觉技术进行刀具磨损状态监测,设计了基于表面微观纹理图像的刀具磨损状态监测实验系统。提出从二维PCA重构图像中提取分形特征值来判断刀具的磨损状态,给出了二维PCA图像重构算法。理论分析和实验证明:PCA重构图像消除了原始图像信息中的冗余和噪声,从重构图像中提取出来的分形布朗运动维数与刀具磨损有着很强的相关性,可以间接判断刀具磨损情况,从而达到对刀具状态进行监测的目的。     

不完备先验知识下的刀具磨损状态评估方法研究

在实际切削加工中刀具磨损的全状态先验知识获取困难,而刀具磨钝状态下的先验知识则较易获取。针对这种不完备先验知识情况,以切削力为监测信号,提出基于连续隐马尔可夫模型(CHMM)的刀具磨损状态评估技术。应用小波包分解技术提取信号特征信息,利用刀具磨钝状态下的先验归一化特征信息建立CHMM监测模型;根据刀具未知状态特性向量与监测模型间的对数似然度获取刀具性能指标,实现刀具磨损状态评价。铣刀全寿命磨损实验表明:该方法能在仅具备磨钝状态先验知识情况下,实现对刀具的磨损状态的初步评估,且所需样本数较少,训练速度快。     

基于特征自适应融合和集成学习的高性能铣削刀具状态监测

通过人工智能、工业大数据实时感知切削加工中的刀具状态是实现面向性能的制造的重要技术途径,也是高性能制造的关键内涵。然而,在目前的切削刀具状态监测算法中,特征提取过程多依赖于人工经验,这无疑限制了刀具状态监测技术的在高性能制造中的推广应用。因此,针对高性能加工监测中的自主性和准确性要求,基于特征自适应融合和集成学习技术,提了出一种面向高性能铣削的刀具磨损监测方法。所提出的监测方法能够根据特征的表现自动为其赋予不同的权重从而实现特征的自适应融合,同时利用AdaBoost集成学习算法,在自动融合特征的同时保证了状态监测精度。薄壁件铣削实验表明,监测结果与真实磨损间的RMSE和MAE值最大为10.44,最小可达5.16。所提出的方法能够自主、准确地监测航空类薄壁件铣削加工中的刀具磨损状态,解决了高性能铣削加工刀具磨损监测中的人工经验依赖问题。     

A microcontroller-based end milling cutter monitoring and management system

The monitoring of end milling cutting operations for tool breakage is achieved using a low-cost microcontroller-based system. The system is based upon acquiring and analysing machine tool-based signals for characteristic responses to tool breakage. Spindle speed and load signals are shown to be responsive to tool condition and thus capable of supporting the deployed approach. The resulting system operates in real time with tool breakage detection consistently diagnosed within two revolutions. The monitoring function is extended to consider tool wear using analysis methods applied in the time and frequency domains. Decisions about tool condition are made by integrating all relevant information into a rule base. Higher-level tool management functions supported by the deployed system are identified.     

介绍一种车床刀具磨损视觉检测装置

对于现代机床而言,刀具的磨损状态监测显得日益重要。在此设计并制作出了一种新颖的在机视觉检测装置,不仅可以实现视觉系统与机床的结合,而且具有良好的隔振性能。实验结果表明,采用这种机构,可以使摄像机拍摄到比较清晰的图像,为实现车刀磨损在机检测提供了条件。     

Dynamic characterization of machining systems

In the working space model of machining, an experimental procedure is implemented to determine the elastic behaviour of the machining system. In this paper, a dynamic characterization and vibration analysis has long been used for the detection and identification of the machine tool condition. The natural frequencies of the lathe machining system are required (Ernault HN400??France) according to three different situations with no cutting process are acquired. The system modal analysis is used to identify the natural frequencies. These frequencies are then compared to the ones obtained on the spindle numerical model by finite element method. This work is validated by experimental tests based on measures of the lathe machine tool frequencies domain. The main objective is to identify a procedure giving the natural frequency values for the machine tool components, in order to establish a better condition in the cutting process of the machine tool.     

Coolant system health monitoring and fault diagnosis via health parameters and fault dictionary

A new methodology is described for the condition monitoring and fault diagnosis of machine tool coolant systems. The steady state characteristics of the coolant system pump outlet pressure, pump motor temperature and tank level are used to define health parameters from which the system health is deduced. On detection of a significant change in system health the pressure transient on closure of the flow valve is captured to aid diagnosis of the fault. A demonstration system for the coolant system of a Wadkin V4-6 machine tool is described. This is used to verify the usefulness of the health system described above by simulation of a number of typical faults.     

An intelligent sensor fusion system for tool monitoring on a machining centre

The objective of this paper is to construct an intelligent sensor fusion monitoring system for tool breakage on a machining centre. Since none of the sensing and diagnosis techniques have proved to be completely reliable in practice, an intelligent tool-monitoring system consisting of a neural-network-based algorithm and a sensor fusion system is proposed. The dual sensing signals of cutting force and acoustic emission are used simultaneously in the proposed system owing to good correlation existing between them, and, a self-learning neural-network algorithm is used to integrate multiple sensing information to make a proper decision about tool condition. The results show good performance in tool-breakage detection by the proposed monitoring system, especially where there is high interference.     

Tool breakage detection in CNC high-speed milling based in feed-motor current signals

Tool condition monitoring, mainly tool breakage detection for high-speed machining (HSM), is an important problem to solve; however, the techniques or types of sensors applied in other research projects present certain inconveniences. In order to improve tool breakage monitoring systems, a simple, effective, and fast method is presented herein. This method is based on the discrete wavelet transform (DWT) and statistical methodologies. The effectiveness of the method is based on the measurements of the feed-motor current signals using inexpensive sensors. It is well-known that during the cutting process, the motor current is related to the tool condition. The current consumption changes when the tool is broken as compared to when the tool is in normal cutting condition. This difference can be obtained from the waveform variances between the signals in order to ascertain the tool condition. The algorithms of this research project consist of obtaining compressed signals from the I _rms feed-motor current signals applying the DWT. Then from these compressed signals, we detect the asymmetries between them. The arithmetic mean value is applied to asymmetries of consecutive machining lengths to reduce noise in the data having a mean value of a series of asymmetries; also, a normal cutting threshold is set up in order to make decisions regarding the tool conditions so as to detect tool breakage. Therefore, this research project shows a low-cost monitoring system that is simple to implement.     

Real‐Time Monitoring of Tool Fracture in Turning Using Sensor Fusion

The sensor fusion method using both an acoustic emission (AE) sensor and a built-in force sensor is introduced for on-line tool condition monitoring during turning. The cutting force was measured by a built-in piezoelectric force sensor, which was inserted in the tool turret housing of an NC lathe. FEM analysis was carried out to locate the most sensitive position for the sensor. A burst of AE signal was used as a triggering signal to inspect the cutting force. A significant drop in cutting force indicated tool breakage. The algorithm was implemented in a DSP board and the monitoring system was installed on a CNC lathe in an FMS line for in-process tool-breakage detection. The proposed system showed an excellent monitoring capability.     

The Application of an ANFIS and Grey System Method in Turning Tool-Failure Detection

The cutting process is a major material removal process; hence, it is important to search for ways of detecting tool failure. This paper describes the results of the application of an adaptive-network-based fuzzy inference system (ANFIS) for tool-failure detection in a single-point turning operation. In a turning operation, wear and failure of the tool are usually monitored by measuring cutting force, load current, vibration, acoustic emission (AE) and temperature. The AE signal and cutting force signal provide useful information concerning the tool-failure condition. Therefore, five input parameters of the combined signals (AE signal and cutting force signal) have been used in the ANFIS model to detect the tool state. In this model, we adopted three different types of membership function for analysis for ANFIS training and compared their differences regarding the accuracy rate of the tool-state detection. The result obtained for the successful classification of tool state with respect to only two classes (normal or failure) is very good. The results also indicate that a triangular MF and a generalised bell MF have a better rate of detection. We also applied grey relational analysis to determine the order of influence of the five cutting parameters on tool-state detection.     

基于小波分析的切削力信号奇异性检测

利用小波变换模的极大值和信号奇异点的关系,分析了用Lip指数来描述的切削力信号局部奇异性.通过观察奇异点的位置等信息得到切削刀具的磨损情况.通过对实际刀具磨损的在线监测数据分析,证明了采用小波变换检测刀具磨损这一方法的有效性.     

A knowledge-based system for sensor interaction and real-time component control

This paper outlines the integration of a two-dimensional vision system with a pneumatic proximity-to-tactile sensing device to form a Co-ordinated ‘Hand-Eye’ system. With the aid of a knowledge base this system is utilised as an intelligent condition monitoring tool for recognition and detection of orientation of parts in a flexible manufacturing environment. An expert system is formulated to interrogate the acquired data streams for the purpose of comparative studies with the knowledge base. Appropriate data processing methods are employed to ensure rapid manipulation of data for real-time applications.