Tool Wear Detection Research Articles

CNN-based in situ tool wear detection: A study on model training and data augmentation in turning inserts

We present an automated turning insert wear detection system developed for aeronautical Low Pressure Turbines (LPT) casing machining process based on a binary classifier using Convolutional Neural Networks (CNN). This method involves acquiring the image on the machine itself. During this process, removing the insert from the tool holder is not necessary, and the wear assessment is performed before the next workpiece is mechanized. Since datasets in tool wear prediction are often imbalanced, a multi perspective camera technology as well as data augmentation and class weighting are utilized to address both the number of worn parts considered and the cost of image acquisition. In this study four different insert types and two CNN architectures (specific and universal models) are considered and evaluated. The effects of data augmentation and training set size are discussed. While the models trained perform well on round inserts, they fail on rhombic insert types. An accuracy up to 97.8% (Matthew’s correlation coefficient of 0.955) is achieved by the machine learning model. Additionally, it can detect defects on a variety of insert types.

Detection of tool wear during machining by designing a novel 12-way 2-shot learning model by applying L2-regularization and image augmentation

Detection of tool wear during machining by designing a novel 12-way 2-shot learning model by applying L2-regularization and image augmentation

Sensor fusion and the application of artificial intelligence to identify tool wear in turning operations

This paper aims to develop an effective sensor fusion model for turning processes for the detection of tool wear. Fusion of sensors’ data combined with novelty detection algorithm and learning vector quantisation (LVQ) neural networks is used to detect tool wear and present diagnostic and prognostic information. To reduce the number of sensors required in the monitoring system and support sensor fusion, the ASPS approach (Automated Sensor and Signal Processing Selection System) is used to select the most appropriate sensors and signal processing methods for the design of the condition monitoring system. The experimental results show that the proposed approach has demonstrated its efficacy in the implementation of an effective solution for the monitoring tool wear in turning. The results prove that the fusion of sensitive sensory characteristic features and the use of AI methods have been successful for the detection and prediction of the tool wear in turning processes and show the capability of the proposed approach to reduce the complexity of the design of condition monitoring systems and the development of a sensor fusion system using a self-learning method.

A novel approach of tool condition monitoring in sustainable machining of Ni alloy with transfer learning models

Cutting tool condition is crucial in metal cutting. In-process tool failures significantly influences the surface roughness, power consumption, and process endurance. Industries are interested in supervisory systems that anticipate the health of the tool. A methodology that utilizes the information to predict problems and to avoid failures must be embraced. In recent years, several machine learning-based predictive modelling strategies for estimating tool wear have been emerged. However, due to intricate tool wear mechanisms, doing so with limited datasets confronts difficulties under varying operating conditions. This article proposes the use of transfer learning technology to detect tool wear, especially flank wear under distinct cutting environments (dry, flood, MQL and cryogenic). In this study, the state of the cutting tool was determined using the pre-trained networks like AlexNet, VGG-16, ResNet, MobileNet, and Inception-V3. The best-performing network was recommended for tool condition monitoring, considering the effects of hyperparameters such as batch size, learning rate, solver, and train-test split ratio. In light of this, the recommended methodology may prove to be highly helpful for classifying and suggesting the suitable cutting conditions, especially under limited data situation. The transfer learning model with Inception-V3 is extremely useful for intelligent machining applications.

Study on In-Situ Tool Wear Detection during Micro End Milling Based on Machine Vision.

Most in situ tool wear monitoring methods during micro end milling rely on signals captured from the machining process to evaluate tool wear behavior; accurate positioning in the tool wear region and direct measurement of the level of wear are difficult to achieve. In this paper, an in situ monitoring system based on machine vision is designed and established to monitor tool wear behavior in micro end milling of titanium alloy Ti6Al4V. Meanwhile, types of tool wear zones during micro end milling are discussed and analyzed to obtain indicators for evaluating wear behavior. Aiming to measure such indicators, this study proposes image processing algorithms. Furthermore, the accuracy and reliability of these algorithms are verified by processing the template image of tool wear gathered during the experiment. Finally, a micro end milling experiment is performed with the verified micro end milling tool and the main wear type of the tool is understood via in-situ tool wear detection. Analyzing the measurement results of evaluation indicators of wear behavior shows the relationship between the level of wear and varying cutting time; it also gives the main influencing reasons that cause the change in each wear evaluation indicator.

Machine Learning in CNC Machining: Best Practices

Building machine learning (ML) tools, or systems, for use in manufacturing environments is a challenge that extends far beyond the understanding of the ML algorithm. Yet, these challenges, outside of the algorithm, are less discussed in literature. Therefore, the purpose of this work is to practically illustrate several best practices, and challenges, discovered while building an ML system to detect tool wear in metal CNC machining. Namely, one should focus on the data infrastructure first; begin modeling with simple models; be cognizant of data leakage; use open-source software; and leverage advances in computational power. The ML system developed in this work is built upon classical ML algorithms and is applied to a real-world manufacturing CNC dataset. The best-performing random forest model on the CNC dataset achieves a true positive rate (sensitivity) of 90.3% and a true negative rate (specificity) of 98.3%. The results are suitable for deployment in a production environment and demonstrate the practicality of the classical ML algorithms and techniques used. The system is also tested on the publicly available UC Berkeley milling dataset. All the code is available online so others can reproduce and learn from the results.

Development of Online Tool Wear-Out Detection System Using Silver-Polyester Thick Film Sensor for Low-Duty Cycle Machining Operations.

This paper deals with the design and development of a silver-polyester thick film sensor and associated system for the wear-out detection of single-point cutting tools for low-duty cycle machining operations. Conventional means of wear-out detection use dynamometers, accelerometers, microphones, acoustic emission sensors, thermal infrared cameras, and machine vision systems that detect tool wear during the process. Direct measurements with optical instruments are accurate but affect the machining process. In this study, the use of a thick film sensor to detect wear-out for aa real-time low-duty machining operation was proposed to eliminate the limitations of the current methods. The proposed sensor monitors the tool condition accurately as the wear acts directly on the sensor, which makes the system simple and more reliable. The effect of tool temperature on the sensor during the machining operation was also studied to determine the displacement/deformation of tracing and the polymer substrate at different service temperatures. The proposed tool wear detection system with the silver-polyester thick film sensor mounted directly on the cutting tool tip proved to be highly capable of detecting the tool wear with good reliability.

Cost-effective classification of tool wear with transfer learning based on tool vibration for hard turning processes

This paper presents a novel application of transfer learning for tool wear detection in turning processes using one-dimensional (1D) convolutional neural network (CNN). The work also investigates the applications of low-cost MEMS as well as IEPE accelerometers in tool wear detection. Tool wear detection is performed using a classification method in which two classes of tool wear sizes are defined: class one and class two which correspond to tool wear sizes smaller or equal to 0.1 mm and bigger than 0.1 mm respectively. The advantage of transfer learning is that it utilizes knowledge from previously learned tasks and applies them to the related ones. In the case of CNC machines, tool wear mechanism in the turning process and the working principles are similar thus, transfer learning can be used to generalize the specific cutting condition to a broader use case. The transfer learning model in this paper uses a pretrained tool wear prediction model which was composed of 1D CNN layers with full connection layers and sufficient amount of data on a source CNC machine. The CNN layers of the pretrained model are frozen at first and then the full connection layers are trained with the new data from the target CNC machine with different cutting insert types. In this study, the application of transfer learning in tool wear size classification showed that the pretrained tool wear detection models can be transferred to other similar processes while maintaining a high accuracy. The accuracy of the transfer learning model was evaluated by comparing it with the results of a model developed from scratch using only the target machine data. It is demonstrated that the transfer learning maintained an accuracy of higher than 80%. In addition to this, the transfer learning model significantly increased the tool wear classification accuracy using the single-axis low-cost MEMS accelerometer from 58% to 85%. Moreover, the tool wear classification model using transfer learning significantly reduced the amount of data required for model development. To evaluate this, the accuracy of the transfer learning model was tested by further reducing the amount of the training data by maximum 80% and the model still showed an accuracy higher or equal to 80%.

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.

Physics-based modeling and information-theoretic sensor and settings selection for tool wear detection in precision machining

Precision machining of metals is an energy intensive process with applications and impacts across the manufacturing industry. The energy efficiency, product yield, and maintenance of the precision machine require a digital twin that can assist with prognostics and health management. Here, a physics-based model is developed and validated against face milling data, and then used for the timely and precise inference of machining faults that cannot be measured directly. Computer numerical control (CNC) measurements of power and force are used through this physics-based machining model to predict deviations of the outputs of power consumption and cutting forces during normal operation. A model-based fault detection and isolation methodology is applied to determine the optimal (traditional and available) sensor suite and the test settings (admissible input values) that improve the inference of tool wear in face milling. The optimal sensor suite and input test settings are obtained by solving a mixed integer non-linear program that optimizes information-theoretic metrics relevant to the detection and isolation of tool wear from steady-state or transient machining measurements. Dynamic time warping and k—NN classification are then used to validate the robustness of the optimal design for fault detection test design, including the optimal sensor suite.

Investigations on the Potential of 5G for the Detection of Wear in Industrial Roller-Burnishing Processes

Roller burnishing represents an economical alternative to conventional surface-finishing processes, such as fine turning or honing. In contrast to the well-known wear mechanisms of chip-forming processes, the wear behavior in roller-burnishing is strongly based on the experience of the machine operators. The nature of the finishing process makes roller-burnishing very sensitive to surface defects, as it is often not possible to rework the last step in a process chain. In the present work, a prototype for a smart roller-burnishing tool with 5G communication is presented, which serves as an inline-monitoring tool to detect tool wear. A suitable metric to monitor the tool wear of the manufacturing roll is suggested, and the potentials of 5G communication for the described use-case are evaluated. Based on the signal-to-noise ratio of the process-force, a metric is found that distinguishes new rolls from worn rolls with very small defects on the micrometer scale. Using the presented approach, it was possible to distinguish the signal-to-noise ratio of a roll with very small wear marks by 3.8% on average. In the case of stronger wear marks, on the order of 20 µm, the difference increased to up to 15.6%.

Acoustic emission signals analysis to differentiate the damage mechanism in the drilling of Al-5%B4C metal matrix composite

Tool wear leads to dimensional inaccuracy and low surface quality in the workpiece, and unexpected sudden tool failure. Detection of tool wear is essential to enhance the quality of manufacturing components and extend tool life. The present work is aimed to investigate the various damage mechanisms involved in the cutting tool and workpiece during drilling of Al-5%B4C composite using acoustic emission technique (AET). The dry drilling experiments were carried out at different spindle speeds and feed rates with high strength steel (HSS) tool. AE time-domain parameters such as count, energy, amplitude and root mean square (RMS) voltage were extracted from the signals and correlated with cutting parameters and tool damage. Fast Fourier transform (FFT) was applied to visualize the frequency components in the AE signals during the drilling process. The wavelet packet transform (WPT) approach was performed to the AE signals to identify and discriminate the various damage mechanism involved in the drilling. The differentiated damage mechanism and their corresponding wavelet energy content were studied. The wavelet energy ratio for decomposed components at different speeds was discussed. The vision measuring microscope was employed to measure the tool wear. The AE features, i.e., AERMS and wavelet coefficient increases with increasing tool wear. A scanning electron microscope was also utilized to characterize the microstructural damage present in the cutting tool and workpiece.

Quantitative evaluation method of tool wear based on morphological characteristics of machined surfaces

Tool wear monitoring techniques have been extensively applied in the modern manufacturing sectors to carry out predictive maintenance and to avoid the massive loss induced by sudden downtime. However, the online tool wear detection based on either the machining vibration or the variation of cutting power always encounters the interference caused by the excessive change of workpiece materials or even by the fluctuations of the coolant pressure. In the present work, a rapid half-online tool wear monitoring method based on the standard sample cutting and focus-variation scanning technique (FVST) was designed to quantify the tool wear extents. Invar alloy was used as the standard sample due to its low thermal expansion and ductile machinability. The results show that the texture of the machined surface is determined by the change of the cutting edge, and thus the surface morphologies can be regarded as the index of tool wear. Tool wear condition was revealed and quantified through the assessment models established based on the characteristic parameters including the width/depth of the texture and the surface roughness of the standard sample machined by new and worn tools under the identical cutting conditions. The results show that the tool wear can be revealed efficiently via the help of the surface morphological analysis, and the method is more none-destructive, rapid, and reliable compared with the conventional tool wear monitoring methods. In sum, the quantitative evaluation method of tool wear based on the morphological characteristics of the machined standard sample surfaces is confirmed capable of improving the accuracy and efficiency of tool wear monitoring.

On-Machine Tool Condition Monitoring System Using Image Processing

An automation system of machine tools can free operators from simple and hard labor by the automatic loading and unloading of workpieces, and allowance of unmanned operation during nights and holidays. To achieve complete shop floor operations, various works that are manually performed by operators need to be automated as well. It is also crucial to accurately monitor tools to detect tool wear and chip winding during machining. In this study, we propose an on-machine tool condition monitoring system that measures tool length and diameter using images to detect tool breakage and winding chips.

Hybrid data-driven and model-informed online tool wear detection in milling machines

Precision machining tool wear is responsible for low product throughput and quality. Monitoring the tool wear online is vital to prevent degradation in machining quality. However, direct real-time tool wear measurement is not practical. This paper presents residual-based anomaly detection models, combining a hybrid model comprised of a physics-based model and a data-driven model (a decision tree or a neural network) to predict signals of interest (e.g., power or forces) under nominal conditions, followed by Page’s cumulative sum test for detecting tool wear on-line using the computer numerical control machine measurements. The most informative features are ranked using dynamic programming and its approximation variants from real-time measurements and machine settings, such as the width of cut, depth of cut, feed rate and spindle speed, that serve as inputs to the predictive models. The baseline nominal model is incrementally updated with experimental data via a gradient boosted adaptation model to generate the residuals that account for discrepancies between the actual machine data under normal conditions and the baseline nominal model predictions. The hybrid model is validated against 20 Mazak milling machine experimental tests and one Haas run-to-failure experiment. The proposed anomaly detector is applied to synthetic data from simulations of the physics-based model at different operating conditions, measurement noise levels, and tool wear levels, and the methods were able to achieve an overall 92% accuracy in data with 1% noise. The anomaly detection methods based on hybrid model reduced the false alarms of either the data-driven or physical-based models alone, and are found to be capable of good online detection of tool wear.

Tool wear monitoring of high-speed broaching process with carbide tools to reduce production errors

Real-time monitoring in CNC machine tools is focused on the early detection of tool wear, and in this way to assess part piece quality. The machining process known as broaching is critical for firtree slots (dovetails) production in turbomachinery components, such as turbine disks. Tight tolerances on one hand, even less than 5 µm in firtree-slots pressure faces, and high productivity on the other, are the two main requirements. Besides, broaching tools are very expensive and the cutting edges wear must be estimated during the process; in fact, tool wear in difficult-to-cut materials machining may cause a waste not only in terms of the tool but also of the very expensive workpieces. Broaching usually is one of the last operations in the process chain, so components start the operation with a very high-added value. Hence, only one bad slot implies an unrecoverable piece and therefore a huge waste of time and money.In this paper, a monitoring method for efficient broaching is proposed by combining real-time monitoring and off-line tool wear inspection. Firstly, the cutting tool characteristics are defined, and those affected by tool degradation. Secondly, some broaching cycles were carried out while measuring a) process accelerations through two accelerometers, b) cutting force by load cells, and c) motor drive consumption. They were simultaneously recorded. Furthermore, the sensitivity between tool wear and broaching process natural frequencies is established.Finally, a series of experimental tests were executed for verification, showing the useful approach for daily life production. The paper focuses on signals and their sensitivity to significant process variations.

Multi-dimension tool wear state assessment criterion on the spiral edge of the milling cutter

In the assessment of the tool wear state for the spiral edge of milling cutter based on machine vision, the traditional assessment criterion is often inaccurate due to the problem of missing of information, especially near the tip area. In order to deal with this problem, different lighting-condition settings and additional non-image information compensation techniques are needed. In view of this, an integration tool wear detection method combined line laser edge detection and machine vision is proposed. Subsequently, the combined data acquirement experimental system is designed and built, which can simultaneously acquire two types of parameters: diameter and images under the same detection condition. Then, based on the above information, a multi-dimension series assessment criterion is proposed consisting of three dimensions index: The one-dimensional assessment index gives the average wear value of the spiral side, which is mainly used as data ordination and characterizes the rule of tool wear degradation; the two-dimensional assessment index describes the contour of the wear region and calculates the area value, which can predict the change of the tool wear stage more precisely; and the three-dimensional assessment index gives the 3D morphology by adding depth information within the wear region and quantifies the volume of the worn-off part, which provides clues for early warning of the deterioration caused by the tool worn. Experiments proved that the multi-dimension series assessment criterion is helpful in reflecting the trend of tool life and giving a more accurate assessment of the tool wear state for the spiral edge of a milling cutter.

Deep learning and rule-based image processing pipeline for automated metal cutting tool wear detection and measurement

Tool wear causes costs and quality problems in metal cutting manufacturing processes. This paper contains an approach of digitalization and big data analytical methods to quantify the wear of metal cutting tools. The method developed consists of a pipeline of deep learning operations for processing tool wear images collected with a digital microscope and is complemented by a rule-based approach to measuring wear along the cutting edge of machining tools. The end-to-end approach allows fully automated tool wear detection and measurement that can be used for inline measurements within CNC machine tools for machining applications.

Prediction of machine tool wear using the weighted k-nearest neighbor method and Neural network model with acceleration response

In the manufacturing industry, with the decrease in the working population, there is a demand for intelligent manufacturing processes (smart factory). When focusing on cutting, predicting and detecting tool wear during machining is an important technical issue for autonomy. However, the tool wear of cutting tools varies from machining to machining even under the same machining conditions, with the same type of tool, and with the same material and shape of workpiece. Therefore, even if a representative wear prediction model is constructed to predict tool wear, the prediction accuracy is low and not practical. In this research, we will verify whether the amount of tool wear can be predicted by a neural network, which is a type of machine learning, using the vibration data obtained during cutting and the past. The accuracy of the proposed method is also verified by comparing the wear volume prediction by one representative neural network with the wear volume prediction by the combination of the proposed multiple neural network.

UTILIZATION OF MACHINE LEARNING APPROACHES FOR TOOL WEAR DETECTION AND PREDICTION IN THE CIRCULAR SAWING PROCESS OF METALLIC MATERIALS

Efficiency plays a major role in many areas of the mechanical processing of metals. Especially in basic provision of raw material, for example by cutting components to length during sawing, high penetration and thus efficiency as well as high robustness of the process is required. Here, the sawing tool and its durability is the decisive component in the process. In today's state of the art, at least in modern sawing machines, the condition is recorded via the sawing tool course or also via the motor current. This gives the operator an initial indication of the tool condition. In current machines an extensive sensory recording for e.g., accelerations as a characteristic for the tool condition is not used. The aim of the investigations described in the paper was to use measurement data from various internal and external sensors to record the state of the sawing tool wear and to analyse it using machine learning approaches. For this purpose, cutting tests were carried out on a modern sawing centre and the acquired measurement data were an-alysed using a convolutional neural network (CNN). During the tests, the internal and external sensors were compared to show which of the sensors used provides the best results in detecting wear on sawing tools. The investigations show that the CNN is suitable for detecting tool wear by means of the sensors used, which can be used for on-line monitoring. Finally, an outlook is given, and it is shown how the CNN can be used for active control in the sawing process.