Online Tool Condition Monitoring Research Articles

Online tool condition monitoring in micromilling using LSTM

Online tool condition monitoring in micromilling using LSTM

Artificial Neural Network on Tool Condition Monitoring in Hard Turning of AISI4140 Steel Using Carbide Tool

Hard turning has replaced conventional grinding in production processes in recent years as an emerging technique. Nowadays, coated carbide tools are replacing expensive CBN inserts in turning. Wear is a significant concern when turning with coated carbide; it immediately affects the acceptability of the machined surface, which causes machine downtime and loss due to wastage in machined parts. Online tool condition monitoring (TCM) is required to prevent such critical conditions. Hard turning differs from conventional turning in energy balance during metal cutting, resulting in greater thrust force; hence, the TCM model presented for conventional turning may not be suitable for hard turning. Hence, tool wear prediction for turning is projected based on thrust force using an artificial neural network (ANN). All of the tests were done using a design of experiments called full factorial design (FFD). The specimens were made of AISI 4140 steel that had been hardened to 47 HRC, and the inserts were made of coated carbide. The most impactful input features for wear, selected based on experimental outputs, were given to the neural network and trained. Tool wear is an estimated output from the training set that has been validated with satisfactory results for random conditions. The 5–10–1 network structure with the Levenberg–Marquardt (LM) learning algorithm, R2 values of 0.996602 and 0.969437 for the training and testing data, and mean square error values of 0.000133152 and 0.004443 for the training and testing data, respectively, gave the best results. The MEP values of 0.575407 and 2.977617 are very low (5%). The LM learning algorithm-based ANN is good at predicting tool wear based on how well it predicts tool wear for both the testing set and the training set.

A novel online tool condition monitoring method for milling titanium alloy with consideration of tool wear law

Due to issues such as limited variability in monitoring data across different tool wear conditions and interference during the machining process, data-driven monitoring models are susceptible to misclassification. Therefore, this paper proposes a pioneering approach that takes into account the tool wear law and the characteristic distribution of tool wear monitoring data. Specifically, the paper proposes an automatic feature extraction and tool condition monitoring method based on Siamese Long Short-term Memory Networks (SLSTMs) to transform the original data distribution, enhance the differentiation of different tool wear condition monitoring data, and achieve accurate prediction of tool wear condition. Additionally, a hybrid data and mechanism-driven tool wear monitoring method is proposed that takes advantage of the fact that tool wear is continuous and irreversible to enable reliable adjustment of the monitoring results of the data-driven model without human intervention. The study conducted milling experiments on a three-axis vertical machining center, using TA2 titanium alloy as the workpiece material. Vibration signals from the spindle in three directions were collected as input to the network, with tool wear state labeled as “0″ or ”1″ as the output. Classical machine learning algorithms such as Decision Trees, K-Nearest Neighbors (KNN) and Support Vector Machines (SVM), as well as classical deep learning algorithms such as Convolutional Neural Networks (CNN), Sparse Stacked Autoencoders (SSAE) and Long Short-Term Memory Neural Networks (LSTM), were used to construct and compare feature extraction and state monitoring models. The proposed model based on Siamese Long Short-Term Memory Neural Networks (SLSTMs) achieved testing average accuracy of 98.2% (σ2 = 0.44), outperforming typical deep learning algorithms such as CNN, LSTM and SSAE as well as traditional machine learning algorithms such as Decision Trees, KNN and SVM in terms of accuracy and robustness. Moreover, the proposed data and mechanism hybrid-driven tool wear monitoring method can effectively improve the monitoring accuracy and robustness of data-driven models such as CNN and SSAE. The monitoring accuracy of CNN and SSAE increased from 95.9% to 97.6% and 91.0% to 94.3%, respectively.

Smart Tool Wear Monitoring of CFRP/CFRP Stack Drilling Using Autoencoders and Memory-Based Neural Networks

The drilling of carbon fiber-reinforced plastic (CFRP) materials is a key process in the aerospace industry, where ensuring high product quality is a critical issue. Low-quality of final products may be caused by the occurrence of drilling-induced defects such as delamination, which can be highly affected by the tool conditions. The abrasive carbon fibers generally produce very fast tool wear with negative effects on the hole quality. This suggests the need to develop a method able to accurately monitor the tool wear development during the drilling process in order to set up optimal tool management strategies. Nowadays, different types of sensors can be employed to acquire relevant signals associated with process variables which are useful to monitor tool wear during drilling. Moreover, the increasing computational capacity of modern computers allows the successful development of procedures based on Artificial Intelligence (AI) techniques for signal processing and decision making aimed at online tool condition monitoring. In this work, an advanced tool condition monitoring method based on the employment of autoencoders and gated recurrent unit (GRU) recurrent neural networks (RNN) is developed and implemented to estimate tool wear in the drilling of CFRP/CFRP stacks. This method exploits the automatic feature extraction capability of autoencoders to obtain relevant features from the sensor signals acquired by a multiple sensor system during the drilling process and the memory abilities of GRU to estimate tool wear based on the extracted sensor signal features. The results obtained with the proposed method are compared with other neural network approaches, such as traditional feedforward neural networks, and considerations are made on the influence that memory-based hyperparameters have on tool wear estimation performance.

Tool condition monitoring based on the fractal analysis of current and cutting force signals during CFRP trimming

Carbon fiber–reinforced plastic (CFRP) is becoming more popular in the aerospace industry due to its high strength-to-weight ratio and low weight. Machining CFRP to achieve the required surface quality, on the other hand, remains a challenge. High temperature in the cutting zone area affects the tool life and surface quality of the machined part. A thermally affected matrix makes an inaccurate interpretation of the surface quality. Then, the roughness parameters cannot be an appropriate indicator for surface evaluation of the CFRP. In the aerospace industry, however, ensuring the acceptable surface quality of a part is essential. Minimizing and controlling tool wear are necessary to avoid degrading the finished surface and losing the dimensional accuracy of the final part. Early detection of tool wear and appropriate surface quality in finishing operations can be achieved using online tool condition monitoring. Cutting forces and electric current signals related to the spindle during machining are very responsive to cutting conditions and can accurately represent tool condition changes. Fractal analysis, as a new approach in the online tool condition monitoring, can assess the tool condition during machining. This research investigates the fractal analysis of the spindle electric current signal and the total cutting force signal while trimming CFRP using a CVD end mill through three different tool life conditions, e.g. new tool, moderately worn tool, and severely worn tool. The empirical fractal index is also introduced to assess the tool condition and ensure the acceptable surface qualities in the finishing operations. The effectiveness of fractal analysis as a decision-making method in the tool condition monitoring was successfully proven in this study.

Overview of Tool Wear Monitoring Methods Based on Convolutional Neural Network

Tool wear monitoring is of great significance for the development of manufacturing systems and intelligent manufacturing. Online tool condition monitoring is a crucial technology for cost reduction, quality improvement, and manufacturing intelligence in modern manufacturing. However, it remains a difficult problem to monitor the status of tools online, in real-time and accurately in the industry. In the research status of mainstream technology, the convolution neural network may be a good solution to this problem, based on the appropriate sensor system and correct signal processing methods. Therefore, this paper outlines the state-of-the-art systems encountered in the open access literature, focusing on information collection, feature selection–extraction technologies based on deep convolutional neural networks, and monitoring network architecture and modeling methods. Based on typical cases, this paper focuses on the application of the convolution neural network in tool wear monitoring. From the application results, it is feasible and reliable to apply convolution neural networks in tool wear monitoring. Additionally, it can improve the prediction accuracy, which is of great significance for the future development of technology. This paper can be a guide for the researchers and manufacturers in the area of tool wear monitoring for explaining the latest trends and requirements.

Transfer of Process References between Machine Tools for Online Tool Condition Monitoring

Process and tool condition monitoring systems are a prerequisite for autonomous production. One approach to monitoring individual parts without complex cutting simulations is the transfer of knowledge among similar monitoring scenarios. This paper introduces a novel monitoring method which transfers monitoring limits for process signals between different machine tools. The method calculates monitoring limits statistically from cutting processes carried out on one or more similar machines. The monitoring algorithm aims to detect general process anomalies online. Experiments comprise face-turning operations at five different lathes, four of which were of the same model. Results include the successful transfer of monitoring limits between machines of the same model for the detection of material anomalies. In comparison to an approach based on dynamic time warping (DTW) and density-based spatial clustering of applications with noise (DBSCAN), the new method showed fewer false alarms and higher detection rates. However, for the transfer between different models of machines, the successful application of the new method is limited. This is predominantly due to limitations of the employed process component isolation and differences between machine models in terms of signal properties as well as execution speed.

Chisel edge wear measurement of high-speed steel twist drills based on machine vision

Tool wear is usually the most relevant parameter of tool performance detection, which directly affects the product quality and tool life. Online tool condition monitoring can effectively reduce workpiece scrap and machine failure. Chisel edge wear is one of the main wear forms in high-speed steel (HSS) twist drill. A machine vision method for chisel edge wear measurement is proposed in this paper to improve measure accuracy and reduce testing costs. Firstly, Otsu segmentation based on Laplacian edge information is proposed for tool image segmentation. A morphology-based Canny operator is developed to perform edge detection and image registration for rough positioning of the tool wear area. Zernike moment-based sub-pixel edge detection is proposed to improve the measurement accuracy. Finally, principal curve is employed to fit sub-pixel edge points to obtain smooth edge curve of the tool wear. Finally, the chisel edge wear can be detected and measured effectively based on the proposed method. In the real machining process, the test results show that the proposed system shows high response speed and inspecting accuracy, which is more convenient and efficient than manual measurement with an electron microscope. This system can be effectively applied to the real-time monitoring of tool wear in industry.

A simple online tool condition monitoring system using artificial neural networks

During machining occurrence of tool wear is a common phenomenon. As tool wear increases, rubbing of flank surface and workpiece also gets increased. Then, the desired quality of workpiece is not possible to achieve. To minimize this loss of quality, a cutting tool should be changed or ground after reaching certain amount of average flank wear (0.3 mm for uniform and 0.6 mm or non-uniform flank wear). For the detection of worn out state of a cutting tool, condition monitoring is required. During past decades, a lot of research works had been done on both offline and online monitoring of cutting tool. Most of the researchers used high cost setups and sensors for wear detection purpose. In this work, tool wear is detected using spindle speed as the wear detection parameter. Artificial Neural Networks is used as data analysing tool. Back propagation algorithm is used as learning algorithm. Results show that proposed methodology is capable to detect tool wear satisfactorily.

A Frontier Statistical Approach Towards Online Tool Condition Monitoring and Optimization for Dry Turning Operation of SAE 1015 Steel

A Frontier Statistical Approach Towards Online Tool Condition Monitoring and Optimization for Dry Turning Operation of SAE 1015 Steel

Online tool condition monitoring for ultrasonic metal welding via sensor fusion and machine learning

In ultrasonic metal welding (UMW), tool wear significantly affects the weld quality and tool maintenance constitutes a substantial part of production cost. Thus, tool condition monitoring (TCM) is crucial for UMW. Despite extensive literature focusing on TCM for other manufacturing processes, limited studies are available on TCM for UMW. Existing TCM methods for UMW require offline high-resolution measurement of tool surface profiles, which leads to undesirable production downtime and delayed decision-making. This paper proposes a completely online TCM system for UMW using sensor fusion and machine learning (ML) techniques. A data acquisition (DAQ) system is designed and implemented to obtain in-situ sensing signals during welding processes. A large feature pool is then extracted from the sensing signals. A subset of features are selected and subsequently used by ML-based classification models. A variety of classification models are trained, validated, and tested using experimental data. The best-performing classification models can achieve close to 100% classification accuracy for both training and test datasets. The proposed TCM system not only provides real-time TCM for UMW but also can support optimal decision-making in tool maintenance. The TCM system can be extended to predict remaining useful life (RUL) of tools and integrated with a controller to adjust welding parameters accordingly.

Tool Wear Condition Monitoring by Combining Variational Mode Decomposition and Ensemble Learning.

Most online tool condition monitoring (TCM) methods easily cause machining interference. To solve this problem, we propose a method based on the analysis of the spindle motor current signal of a machine tool. Firstly, cutting experiments under multi-conditions were carried out at a Fanuc vertical machining center, using the Fanuc Servo Guide software to obtain the spindle motor current data of the built-in current sensor of the machine tool, which can not only apply to the actual processing conditions but, also, save costs. Secondly, we propose the variational mode decomposition (VMD) algorithm for feature extraction, which can describe the tool conditions under different cutting conditions due to its excellent performance in processing the nonstationary current signal. In contrast with the popular wavelet packet decomposition (WPD) method, the VMD method was verified as a more effective signal-processing technique according to the experimental results. Thirdly, the most indicative features that relate to the tool condition were fed into the ensemble learning (EL) classifier to establish a nonlinear mapping relationship between the features and the tool wear level. Compared with existing TCM methods based on current sensor signals, the operation process and experimental results show that using the proposed method for the monitoring signal acquisition is suitable for the actual processing conditions, and the established tool wear prediction model has better performance in both accuracy and robustness due to its good generalization capability.

Fractal analysis implementation for tool wear monitoring based on cutting force signals during CFRP/titanium stack machining

Hybrid structures of metals and composite materials are increasingly common in aerospace industry, and the optimization and monitoring of the machining of these stacks are an area of active research. Online tool condition monitoring in particular is a valuable capability and is facilitated by real-time treatment of cutting force signals. Cutting force signals are considered one of the most important measures for tool condition monitoring. The present work treats online cutting force time series with fractal analysis. The signal features generated are central to tool wear assessment. This work evaluates the fractal dimension of the cutting force signals from orbital drilling of a stack of carbon fiber–reinforced plastics (CFRP) and Ti6Al4V titanium alloy as a measure of the “roughness” of these signals. It is shown that distinct wear stages are adequately identified using fractal signal features. Low machining quality may thereby be prevented. Additionally, to address the inconvenient need for long machining tests when studying the application of these techniques to CFRP and titanium alloys, a novel fractal index is proposed to improve the monitoring process without requiring extensive experimentation.

Application of spindle power signals in tool condition monitoring based on HHT algorithm

In the present study, we present the application of spindle power signals in tool condition monitoring (TCM) under different cutting conditions based on the Hilbert-Huang transform (HHT) algorithm. We extracted two features from the original collected data using the HHT algorithm to detect the tool wear and conducted six sets of cutting experiments to verify the feasibility of this tool condition monitoring method. The results show that these features are highly correlated with the wear state of cutting tools, regardless of the cutting parameters, workpiece materials, and machining methods. The calculated correlation coefficients between the extracted features and the actual tool wear reach 0.79–0.98. This demonstrates that the HHT algorithm is suitable for extracting features from the spindle power signals to construct the online tool condition monitoring system.

Online Tool Condition Monitoring Based on Parsimonious Ensemble.

Existing methodologies for tool condition monitoring (TCM) still rely on batch approaches which cannot cope with a fast sampling rate of a metal cutting process. Furthermore, they require a retraining process to be completed from scratch when dealing with a new set of machining parameters. This paper presents an online TCM approach based on Parsimonious Ensemble+ (pENsemble+). The unique feature of pENsemble+ lies in its highly flexible principle where both the ensemble structure and base-classifier structure can automatically grow and shrink on the fly based on the characteristics of data streams. Moreover, the online feature selection scenario is integrated to actively sample relevant input attributes. This paper presents advancement of a newly developed ensemble learning algorithm, pENsemble, where the online active learning scenario is incorporated to reduce the operator's labeling effort. The ensemble merging scenario is proposed which allows reduction of ensemble complexity while retaining its diversity. Experimental studies utilizing two real-world manufacturing data streams: 1) metal turning and 2) 3-D-printing processes and comparisons with well-known algorithms were carried out. Furthermore, the efficacy of pENsemble+ was examined using benchmark concept drift data streams. It has been found that pENsemble+ incurs low structural complexity and results in a significant reduction of the operator's labeling effort.

Tool condition monitoring using spectral subtraction and convolutional neural networks in milling process

Process monitoring is necessary in machining operation to increase productivity, improve surface quality, and reduce unscheduled downtime. Tool wear and breakage are important and common source of machining problems due to high temperatures and forces of the machining process. Therefore, it is highly beneficial to develop an online tool condition monitoring (TCM) system. This paper investigates a robust tool wear monitoring system for milling operation. Recent developments in machine learning, in particular deep learning methods, result in significant improvement in automation of different industries. Therefore, in this research, we employed convolutional neural network (CNN) as a well-established and powerful deep learning algorithm for tool wear estimation. Wavelet packet-based features are extracted for tool wear monitoring as a powerful time-frequency fault indicator. Moreover, a hybrid feature extraction method is proposed using wavelet time-frequency transformation and spectral subtraction algorithms to intensify the effect of tool wear in the signal and reduce the effect of other cutting parameters. CNN-based monitoring systems are compared with three other machine learning methods (support vector machine, Bayesian rigid network, and K nearest neighbor method) as the baseline. The research is validated using different datasets. The algorithms are implemented and compared using experimental force and vibration signals from LIPPS lab of ETS university as well as using current signals as the fault indicator from Nasa_Ames dataset.

Machine learning based tool condition classification using acoustic emission and vibration data in high speed milling process using wavelet features

It is important to develop an intelligent tool condition monitoring system to increase productivity and promoting automation in metal cutting process. Many attempts have been made in the past to develop such systems using signals from various sensors such as dynamometer, current, accelerometer, aco ustic emission, current and voltage, etc. But the successes of different sensor based systems are limited due to the complexity of tool wear process. The research is still ongoing for improved tool condition monitoring system with applications of advance signal processing techniques and artificial intelligent models. In this study, tool conditions are monitored using the vibration and acoustic emission signatures during high speed machining of titanium alloy (Ti-6Al-4V). Using discrete wavelet transforms wavelets coefficients of vibration and acoustic emission signals are extracted using haar, daubechies, biorthogonal and reverse biorthogonal wavelets. Machine learning algorithms such as decision tree, naive bayes, support vector machine and artificial neural networks are used to predict the tool condition. Results indicate the effectiveness of acoustic emission and vibration data using wavelets for classifying the tool conditions with the aid of machine learning algorithms. A correlation is established between the tool conditions and sensor data. Support vector machine trained by vibration data appears to be predicting the tool conditions with good accuracy compared to decision trees, naive bayes and artificial neural network. Results obtained in this study will be useful to develop an intelligent on-line tool condition monitoring system.

A multi-sensor based online tool condition monitoring system for milling process

Tool condition monitoring has been considered as one of the key enabling technologies for manufacturing optimization. Due to the high cost and limited system openness, the relevant developed systems have not been widely adopted by industries, especially Small and Medium-sized Enterprises. In this research, a cost-effective, wireless communication enabled, multi-sensor based tool condition monitoring system has been developed. Various sensor data, such as vibration, cutting force and power data, as well as actual machining parameters, have been collected to support efficient tool condition monitoring and life estimation. The effectiveness of the developed system has been validated via machining cases. The system can be extended to wide manufacturing applications.

A machine vision system for micro-milling tool condition monitoring

Tool condition monitoring is a key issue in micromachining for part quality control because the excessive tool wear and abnormal tool conditions will significantly decrease the size accuracy of part and shorten the tool durability as well. In view of this, a novel configuration of machine vision system for online tool condition monitoring is presented to improve the part quality and extend the micro tool life. The vision system is committed to automated on-machine vision inspection for monitoring the progressive wear. This inspection system uses a telecentric lens with light source and a camera to minimize the errors in imaging. The control system drives a three dimensional motion platform carrying the imaging device to probe and grab in-focus image at the predetermined time interval of machining. In addition to the flank wear, three new wear variables are explored to enhance the robustness in prediction of tool wear state. Effective image processing algorithms are developed to reduce downtime. The effectiveness of the prototype system and the developed algorithms for tool wear extraction are verified by cutting experiments using two-flutter micro-milling tools, and the experimental results show that this novel on-machine vision inspection system is convenient and effective to measure the amount of progressive wear and reflect the trend of tool life.

Development and analysis of an online tool condition monitoring and diagnosis system for a milling process and its real-time implementation

This paper addresses the development of an online tool condition monitoring and diagnosis system for a milling process. To establish a tool condition monitoring and diagnosis system, three modeling algorithms–an Adaptive neuro fuzzy inference system (ANFIS), a Back-propagation neural network (BPNN) and a Response surface methodology (RSM)–are considered. In the course of modeling, the measured milling force signals are processed, and critical features such as Root mean square (RMS) values and node energies are extracted. The RMS values are input variables for the models based on ANFIS and RSM, and the node energies are those for the BPNN-based model. The output variable is the confidence value, which indicates the tool condition states–initial, workable and dull. The tool condition states are defined based on the measured flank wear values of the endmills. During training of the models, numerical confidence values are assigned to each tool condition state: 0 for the initial, 0.5 for the workable and 1 for the dull. An experimental validation was conducted for all three models, and it was found that the RSM-based model is best in terms of lowest root mean square error and highest diagnosis accuracy. Finally, the RSM-based model was used to build an online system to monitor and diagnose the tool condition in the milling process in a real-time manner, and its applicability was successfully demonstrated.