数控机床刀具磨损监测实验数据处理方法研究

数控机床刀具磨损监测对于提高数控机床利用率,减小由于刀具破损而造成的经济损失具有重要意义.有针对性地回顾了国内外各种分析刀具磨损信号方法的研究工作,详细叙述了功率谱分析法、小波变换、人工神经网络以及多传感器信息融合技术的实现形式.通过比较各种数据处理方法的优缺点,提出基于混合智能多传感器信息融合技术是数控机床刀具磨损监测实验数据处理的未来发展的主要方向.     

数控机床刀具磨损监测方法研究

数控机床刀具磨损监测对于提高数控机床利用率,减小由于刀具破损而造成的经济损失具有重要意义.文章有针对性地回顾了国内外各种刀具磨损监测方法的研究工作,详细叙述了切削力监测法、切削噪声监测法、功率监测法、声发射监测法、电流监测法以及基于多传感器监测法等六种刀具磨损监测方法.本文通过比较各种监测方法的优缺点,提出基于多传感器监测法是数控机床刀具磨损监测方法的未来发展的主要方向.     

Identification of tool wear using acoustic emission signal and machine learning methods

The work concerns the monitoring of the edge condition based on acoustic emission (AE) signals. The tool edge condition was determined by the wear width on the flank face. The processed material was an aluminum-ceramic composite containing 10% SiC. A carbide milling cutter with a diamond coating was used as the tool. Based on the AE signals, appropriate measures were developed that were correlated with the edge condition. Machine learning methods were used to assess the milling cutter's degree of wear based on AE signals. The applied approach using a decision tree allowed the prediction error of the tool condition class with a value below 6%. The method was also compared with other machine learning methods such as neural networks and the k-nearest neighbor algorithm.     

A method for detecting tool wear on a CNC lathe using a doppler radar detector

Installing a non-contact in-process tool wear detection system on a computer numerical control lathe can help prevent product defects and improve product quality without impacting product cycle time. Many methods have been proposed for non-contact in-process tool wear detection. In particular, a recent international patent application describes a method for measuring the torque in a rotating axle using a high-frequency wireless transmitter/receiver and a vibrating string. The method has reportedly been used to detect cutting on a manual lathe. The authors present a new method for measuring tool wear using a high-frequency wireless transmitter/receiver alone, without a vibrating string. The high-frequency transmitter/receiver apparently responds to metal-metal contact noise rather than, or more strongly than, to signals generated by a vibrating string. The findings could help bring automated tool wear monitoring systems closer to the level of performance needed for practical use in industry.     

应用PC和80C196KC单片机的刀具磨损状态监测系统

设计一种融合声发射(AE)、主轴电动机电流和Z向进给电动机电流多特征参数检测方法的、以PC机为上位机、以80C196KC单片机为下位机的刀具磨损监测系统。主要介绍系统的硬件结构,阐述系统中多路信号采集装置硬、软件工作原理与设计中的关键技术,以及具有辨识功能的上位数据处理计算机的软件工作流程。     

Active features extracted by deep belief network for process monitoring

Recently, based on the powerful capability of feature extraction, deep learning technique has been applied to the field of process monitoring, and usually, the researches utilize all the abstract features to establish the detection model and detect or classify the fault. However, whether all the extracted features are valid and beneficial for process monitoring have never been researched and discussed. If there are some features that are adverse for process monitoring, the detection performance of the model would be reduced once they are considered in the model, and utilized the features that are advantageous for process monitoring could ameliorate the performance of detection model. Motivated by this, a feasibility analysis on each feature captured by deep belief network for process monitoring is executed and the conception of active features (AFs) which have active expression for the occurrence of the fault is proposed. Based on AFs, utilized Euclidean metric to calculate the dissimilarity between the test sample and the training sample, and moving average technique is employed to reduce the effect of the burst noise in measurement variables on the result. Finally, the comparison of fault detection rate with other advanced methods on a numerical process and TE process demonstrate the feasibility and superiority of the proposed method, AF-DBN in this study.     

Fuzzy c-means clustering based colour image segmentation for tool wear monitoring in micro-milling

Micro-milling is an extensively used micro-machining process for producing high precision 3D components from varied materials. However, tool wear in micro-tools is a big concern, as component accuracy directly depends on it. Also, size effects limit the monitoring by the naked eye, but it can be compensated by implying a proper wear monitoring mechanism. Various direct and indirect methods have earlier been used for monitoring purposes, and considering the needs of the fourth industrial revolution, one of the direct methods, machine vision, when combined with image processing algorithms, can play a more prominent role. Current work focuses on creating a wear monitoring algorithm based on fuzzy c-means clustering technique directly implied on acquired colour micro-tool images. The proposed algorithm has three steps: the first step is Region of Interest (ROI) extraction, where the background is removed, orientation correction is done, and ROI on each tooth is extracted from micro-tool colour images. The second uses the fuzzy c-means technique on ROI to cluster them, from which wear cluster is chosen and morphologically enhanced. The last step performs pixel level measurement and results in numerical wear width. Overall, quantitative results at each step are correlation coefficient of 99 % after image registration, segmentation accuracy of 92 % and wear measurement accuracy of 97 %. A comparison is also made between the proposed algorithm, k-means clustering and RGB thresholding technique, where the proposed algorithm outshines. Lastly, the wear measurement error of the proposed algorithm is less than 5 %, indicating its repeatable, reliable, and robust nature.     

基于复映测量的球头刀具磨损数据处理及其软件开发

在铣削加工中,刀具磨损对产品加工质量有重要的影响,以球头刀具磨损为研究对象,为了建立刀具磨损模型,采用复映测量手段获取刀具磨损,在复映测量获得刀具磨损的过程,针对复印测量的球头铣削刀具磨损数据提出了数据处理的方法,并以Visual C++为开发平台实现了基于复映测量的刀具磨损数据处理模块,为基于复映测量的刀具磨损建模提供了基础,所提出的刀具磨损数据处理方法可有效对刀具磨损数据进行分析建立刀具磨损模型。     

An in-depth study of tool wear monitoring technique based on image segmentation and texture analysis

We present a new micro-vision system for tool wear monitoring, which is essential for intelligent manufacturing. The tool wear area is divided into regions by a watershed transform, then subjected to automatic focusing and segmentation. The individual pixel gray values in each region are then replaced with the corresponding regional mean gray value. A hill climbing algorithm based on the sum modified laplacian (SML) focusing evaluation function is used to search the focal plane. In addition, we implement an adaptive Markov Random Field (MRF) algorithm to segment each region of tool wear. For our MRF model, the connection parameter value is adaptively determined by the connection degree between regions, which improves image acquisition of more integral tool wear areas. Our findings suggest that automatic focusing and segmentation of the tool wear area by region (within the tool wear area) enhance accuracy and robustness, and allow for real time acquisition of tool wear images. We also implement a complementary tool wear assessment procedure based on the surface texture of the workpiece. The optimal texture analysis window is determined using the entropy metric – a texture feature generated using a Gray Level Co-occurrence Matrix (GLCM). In the best texture analysis window, entropy remains monotonic as tool wear increases, demonstrating that entropy can be used effectively to monitor tool wear. Information from combined measurements of tool wear and workpiece texture can reliably be used to monitor tool wear conditions and improve monitoring success rates.     

Adaptive modelling of the milling process and application of a neural network for tool wear monitoring

An adaptive signal processing scheme that uses a low-order autoregressive time series model is introduced to model the cutting force data for tool wear monitoring during face milling. The modelling scheme is implemented using an RLS (recursive least square) method to update the model parameters adaptively at each sampling instant. Experiments indicate that AR model parameters are good features for monitoring tool wear, thus tool wear can be detected by monitoring the evolution of the AR parameters during the milling process. The capability of tool wear monitoring is demonstrated with the application of a neural network. As a result, the neural network classifier combined with the suggested adaptive signal processing scheme is shown to be quite suitable for in-process tool wear monitoring     

基于小波分析和模糊控制理论的铣刀磨损监控方法

研制了一种铣刀磨损的监控方法.在该系统中信号采集采用声发射传感器,信号的特征提取采用小波分析的方法,将变换后的尺度系数和各个频段的小波系数作为特征,采用自行设计的Sugeno模糊控制系统进行状态识别,模糊控制系统的输出是刀具磨损的具体值.     

An automatic system based on vibratory analysis for cutting tool wear monitoring

The aim of this work is to develop a new, simple to use and reliable automatic method for detection and monitoring wear on the cutting tool. To achieve this purpose, the vibratory signatures produced during a turning process were measured by using a three-axis accelerometer. Then, the mean power analysis was proposed to extract an indicator parameter from the vibratory responses, to be able to describe the state of the cutting tool over its lifespan. Finally, an automatic detector was proposed to evaluate and monitor tool wear in real time. This detector is efficient, simple to operate in an industrial environment and does not require any protracted computing time.     

基于光栅投影技术的刀具磨损三维特征提取方法

提出了利用激光光栅投影技术对硬合金刀具磨损图像进行检测的方法。利用采集待测物体栅线图样中未变形区域包含的相位信息,使用最小二乘迭代法将数据拟合,得到了理想参考面相位分布参数。由于不需另外采集实物参考面变形光栅图像,不仅减少了实物参考面的采集步骤,而且不需要对实物参考面相位进行求解。实验表明:该方法利用数控机床换刀间隙进行检测,实现了在线测量。同时,该方法测量磨损区域的最大宽度、最大面积、磨损区域的周长、z方向最大深度参数的标准差和相对精度都小于0.01,具有较好的三维形貌轮廓检测精度。     

基于动态统计滤波与深度学习的智能故障诊断方法

电流信号具有易采集、不易受环境噪声影响的优点,为难以通过振动传感器采集信号的特殊设备提供了可行的监测诊断思路,但电流信号也存在故障特征难以提取等问题。为此,将改进的动态统计滤波与深度卷积神经网络(DCNN)结合,提出一种基于电流信号进行机械设备智能故障诊断的方法。引入综合信息量指标(SIpq)优化滤波效果,基于改进的动态统计滤波方法,使不同状态信号间的特征差异最大化,以提高状态识别精度;通过交替堆叠特征图尺寸不变的卷积层与逐层递减的池化层,构建DCNN,提取电流信号中的高维故障特征。将动态统计滤波后的特征增强图像输入DCNN,识别故障类型。为验证方法有效性,以不平衡、不对中、松动3种故障为对象进行故障类型识别,分析结果表明,所提方法可有效识别故障类型,与传统的ANN、CNN等其他方法对比具有较好的识别精度。     

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。     

基于深度学习的电机轴承微小故障智能诊断方法*

运用深度学习技术对滚动轴承微小故障发生的位置、类别和严重程度进行精准自动的辨识是当前故障诊断领域研究的热点。传统的故障诊断方法过度依赖于工程师凭经验进行手工特征提取,难以有效提取微小故障特征。提出了一种改进的CNNs-SVM的新方法用于电机轴承的故障快速智能诊断,该方法采用1×1的过渡卷积层与全局均值池化层的组合代替传统CNN的全连接网络层结构,有效减少CNN的训练参数量,在测试阶段采用支持向量机代替Softmax分类器进一步提升诊断准确率。最后将提出的方法用于电机支撑滚珠轴承的故障实验数据并与多种算法对比验证。结果表明,改进CNNs-SVM算法的故障识别准确率高达99.86%,同时在不同负载下具有良好的迁移泛化能力,具备实际工程应用的可行性。其诊断准确率和测试时间明显优于其他智能算法。     

多传感器融合下多工况刀具磨损状态预测的深度森林方法研究

准确监测加工过程刀具磨损状态有助于避免因刀具失效导致的产品质量问题。 建立不同工况的刀具磨损监测模型,往往需要对每组工况调参以保证精度。 为减少调参并保证预测精度,结合深度森林的超参数少、参数对模型不敏感和训练过程自适应等优点,利用深度森林建立了多传感器信号及多工况下自主特征选择的刀具磨损状态预测模型。 基于 3 组不同工艺参数下 TC18 铣削过程的多传感器及磨损数据,以及预测与健康管理(PHM)学会 2010 年高速数控机床刀具健康预测竞赛的开放数据,深度森林在 3 组工况的预测精度分别为 95. 35% 、96. 63% 和 97. 06% ,在 PHM 数据上为 98. 95% ,验证了深度森林对多工况下刀具磨损预测的高精度和适用性,为在线监测技术提供了有力的指导。     

基于DBN的故障特征提取及诊断方法研究

随着装备日趋复杂化,依靠专家经验或信号处理技术人工提取和选择故障特征变得越来越困难。此外,以BP神经网络、SVM为代表的浅层模型难以表征被测信号与装备健康状况之间复杂的映射关系,且面临维数灾难等问题。结合深度置信网络(DBN)在提取特征和处理高维、非线性数据等方面的优势,提出一种基于深度置信网络的故障特征提取及诊断方法。该方法通过深度学习利用原始时域信号训练深度置信网络并完成智能诊断,其优势在于能够摆脱对大量信号处理技术与诊断经验的依赖,完成故障特征的自适应提取与健康状况的智能诊断,该方法对时域信号没有周期性要求,具有较强的通用性和适应性。在仿真数据集和轴承数据集上进行了故障特征提取和诊断实验,实验结果表明:本文提出的方法能够有效地从原始信号中进行多种工况、多种故障位置和多种故障程度的故障特征提取和诊断,并且具有较高的故障识别精度。     

基于自注意力机制的深度学习模拟电路故障诊断

模拟电路是集成电路中的重要组成部分,基于深度学习技术对模拟电路发生的故障进行检测,并精准识别故障的类型是当前集成电路测试领域的研究热点。针对模拟集成电路故障检测存在困难的问题,利用人工智能在图像识别领域、语音分类领域的先进技术,提出了基于自注意力机制检测Sallen-Key型低通滤波电路故障的深度学习模拟电路故障检测方案,将输出信号采样成音频信号,并将其输入到自注意力变换网络的音频分类模型中进行训练、测试和优化。结果表明,通过自注意力变换网络音频分类在9种不同的故障类型诊断中,平均准确率达93.1%,最高准确率达98.1%。该模型收敛速度更快,具有较强的模拟电路故障检测能力。     

A new automatic convolutional neural network based on deep reinforcement learning for fault diagnosis

Convolutional neural network (CNN) has achieved remarkable applications in fault diagnosis. However, the tuning aiming at obtaining the well-trained CNN model is mainly manual search. Tuning requires considerable experiences on the knowledge on CNN training and fault diagnosis, and is always time consuming and labor intensive, making the automatic hyper parameter optimization (HPO) of CNN models essential. To solve this problem, this paper proposes a novel automatic CNN (ACNN) for fault diagnosis, which can automatically tune its three key hyper parameters, namely, learning rate, batch size, and L2-regulation. First, a new deep reinforcement learning (DRL) is developed, and it constructs an agent aiming at controlling these three hyper parameters along with the training of CNN models online. Second, a new structure of DRL is designed by combining deep deterministic policy gradient and long short-term memory, which takes the training loss of CNN models as its input and can output the adjustment on these three hyper parameters. Third, a new training method for ACNN is designed to enhance its stability. Two famous bearing datasets are selected to evaluate the performance of ACNN. It is compared with four commonly used HPO methods, namely, random search, Bayesian optimization, tree Parzen estimator, and sequential model-based algorithm configuration. ACNN is also compared with other published machine learning (ML) and deep learning (DL) methods. The results show that ACNN outperforms these HPO and ML/DL methods, validating its potential in fault diagnosis.