Tool Wear Condition Research Articles

Physics-informed hidden markov model for tool wear monitoring

Tool wear monitoring (TWM) is an important part of modern intelligent machining. An efficient and accurate monitoring system will effectively improve machining accuracy and reduce processing costs. At present, the methods for TWM can generally be divided into three categories: physics-based models, data-driven models, and hybrid models. These current approaches are suffered from either low prediction accuracy, insufficient model training, or poor result interpretability due to the nature of the model construction approaches. To overcome the shortcomings of these methods, a physics-informed hidden Markov model (PI-HMM) proposed in this study fully integrates the advantages of the above two methods. Firstly, the model constrains the division of hidden states through the tool wear physical model to improve the physical consistency of the model. Secondly, according to the process parameters and the labels generated by the tool wear physical model, the training set is expanded to solve the problem of insufficient training data of the data model. Finally, the influence of real-time features is added to the established wear state output model and the physical data fusion training strategy is used to train the model to realize the classification of tool wear state and the regression of wear value. The experimental results show that the proposed model in this study achieves an average recognition rate of 0.995 for wear classification and an average coefficient of determination (R2) of 0.968 for wear regression.

Tooth-wise monitoring of the asymmetrical tool wear in micro-milling based on the chip thickness reconstruction and cutting force signal

Tooth-wise monitoring of the tool wear condition is of great importance for improving the machining quality and efficiency of micro-milling. However, the uncut chip thickness, cutting force, and tool wear condition of the multi-tooth evolve asymmetrically coupled due to the tool runout, and it is difficult to monitor the tool wear of each cutting tooth solely based on the cutting force signal. The reconstruction of chip thickness is the key to overcoming this challenge. With such consideration, this study proposes a tooth-wise monitoring method for estimating the asymmetrical micro-milling tool wear based on the chip thickness reconstruction and cutting force signal. The instantaneous uncut chip thickness of each tooth is reconstructed from the floor surface topography of the machined material with the guide of the proposed surface topography model considering the asymmetrical tool wear. The wear condition of each tooth is estimated by integrating the reconstructed chip thickness and the measured cutting force signal into the proposed asymmetrical micro-milling force model. Experimental results show that the asymmetrical models outperform the traditional symmetrical models in terms of modeling flexibility and accuracy. With the consideration of the asymmetrical tool wear, the prediction accuracy of the proposed topography and cutting force models increases by 11.4% and 5.3%. The relative monitoring error within 9% demonstrates that the proposed method can accurately estimate the tool wear condition of each cutting tooth in micro-milling.

On the tool wear mechanism of machining Zr-based bulk metallic glasses under varying corner radii

Revealing the tool wear mechanisms of machining Zr-based bulk metallic glasses (BMGs) is important for the processing of BMGs in engineering applications. The tool wear mechanisms of machining BMGs with PCBN tools under varying corner radii were investigated, and the results shown that the adhesive and oxidation wear were the main wear mechanisms. The low thermal conductivity of the BMG workpiece material increased the cutting temperature. The effect of corner radius on tool wear mechanism was characterized by the normal stress fields. It indicated that the extremely high maximum normal stress was the main reason for edge chipping. Finally, the influence of tool wear state on machined surface quality was related to the machined surface morphology. The findings have shown that the edge chipping and adhesive wear produced feed marks, material side flow and adhensions on the machined surface, leading to the increase of surface roughness.

Image super resolution by double dictionary learning and its application to tool wear monitoring in micro milling

It is an effective means to improve the machining quality of product by monitoring the tool wear conditions in micro milling. However, due to the small size and high speed of the micro milling cutter, it is difficult to capture the clear image of tool wear by CCD camera. Therefore, a double dictionary sparse coding super-resolution (SR) method is proposed to improve the resolution of micro milling tool wear images in this work. Based on the traditional single dictionary, this work proposes the residual dictionary to reconstruct the residual image of the tool, achieving the double dictionary coefficient encoding reconstruction. In order to extract local features of low resolution (LR) tool wear images, the Gobor filter banks are designed, and then the PCA method is used to reduce dimension of extracted features. The dictionary learning based on a single high resolution (HR) image is realized, which reduces the dependence of super-resolution algorithm on large datasets. The effectiveness of Gabor filter and double dictionary for restoration tool wear image details is verified by experiments. Compared with proposed method, traditional interpolation method, sparse coding super resolution model [32] and super resolution convolutional neural network model [26], the PSNR mean value of those methods are 43.62, 42.19, 43.20, 43.21, respectively, which shows that the proposed method has excellent performance in super-resolution reconstruction of micro milling tool images for tool wear monitoring.

Study on tool wear state recognition algorithm based on spindle vibration signals collected by homemade tool condition monitoring ring

With a view to further realising the intelligence of tool condition monitoring (TCM) and to address the high cost and low stiffness problems of existing smart tool holders, this study firstly proposed a tool condition monitoring ring, which can be mounted on the tool shank and is easily disassembled and reused. This ring acquires the triaxial vibration signals of the spindle. This study also firstly applied manifold learning to TCM model for data dimensionality reduction and proposed a weighted random forest based on K-nearest neighbour (KWRF). A tool wear state recognition algorithm based on wavelet threshold de-noising (WTD), variational mode decomposition (VMD), manifold learning and KWRF was proposed. Four sets of full life cycle milling experiments were carried out on 45 steel with cemented carbide milling inserts. The results showed that the recognition accuracy of the proposed model reached 98.74%, which verified the effectiveness of the proposed model.

Research on tool condition monitoring (TCM) using a novel unsupervised deep neural network (DNN)

In order to improve the recognition precision and accuracy of tool wear monitoring, an unsupervised deep neural network (DNN) based on stack denoising autoencoder (SDA) is proposed. After feature extraction and selection, the stack denoising automatic coding network reduces the dimensionality of the feature vector. On this basis, principal component analysis (PCA) and T-distributed random neighbor embedding (t-SNE) are used to reduce the dimensionality of the features twice, and finally a simple two-dimensional feature matrix is obtained. Finally, the deep neural network model of SDA is established by adding SoftMax regression layer, and the tool wear monitoring results are taken as new labeled data, and the deep neural network parameters are fine-tuned by secondary backpropagation. The experimental results show that the proposed method can learn adaptively and obtain effective feature expression, and the tool wear state recognition results are highly accurate. The proposed method can effectively identify the tool wear state.

Tool Wear State Recognition Based on One-Dimensional Convolutional Channel Attention.

Tool wear state recognition is an important part of tool condition monitoring (TCM). Online tool wear monitoring can avoid wasteful early tool changes and degraded workpiece quality due to later tool changes. This study incorporated an attention mechanism implemented by one-dimensional convolution in a convolutional neural network for improving the performance of the tool wear recognition model (1DCCA-CNN). The raw multichannel cutting signals were first preprocessed and three time-domain features were extracted to form a new time-domain sequence. CNN was used for deep feature extraction of temporal sequences. A novel 1DCNN-based channel attention mechanism was proposed to weigh the channel dimensions of deep features to enhance important feature channels and capture key features. Compared with the traditional squeeze excitation attention mechanism, 1DCNN can enhance the information interaction between channels. The performance of the model was validated on the PHM2010 public cutting dataset. The excellent performance of the proposed 1DCCA-CNN was verified by the improvement of 4% and 5% compared to the highest level of existing research results on T1 and T3 datasets, respectively.

Tool Wear State Identification Based on SVM Optimized by the Improved Northern Goshawk Optimization.

Tool wear condition significantly influences equipment downtime and machining precision, necessitating the exploration of a more accurate tool wear state identification technique. In this paper, the wavelet packet thresholding denoising method is used to process the acquired multi-source signals and extract several signal features. The set of features most relevant to the tool wear state is screened out by the support vector machine recursive feature elimination (SVM-RFE). Utilizing these selected features, we propose a tool wear state identification model, which utilizes an improved northern goshawk optimization (INGO) algorithm to optimize the support vector machine (SVM), hereby referred to as INGO-SVM. The simulation tests reveal that INGO demonstrates superior convergence efficacy and stability. Furthermore, a milling wear experiment confirms that this approach outperforms five other methods in terms of recognition accuracy, achieving a remarkable accuracy rate of 97.9%.

The machinability of titanium alloy thin-wall parts in cooling minimum quantity lubrication (CMQL) environments

The machining of thin-wall components made of titanium alloys is challenging because the poor machinability of the material leads to severe problems such as accelerated tool wear and poor surface quality, while the weak rigidity of the thin-wall structure results in unavoidable vibration and surface form errors. To address these issues, this paper investigated the mechanisms and performance of cooling minimum quantity lubrication (CMQL) in milling titanium thin-wall parts. To verify the efficiency of CMQL, different cooling/lubrication strategies, including conventional flood cooling, minimum quantity lubrication (MQL) and CMQL with different temperature levels, were investigated. The cutting force, tool wear state, chip formation, surface integrity, and surface form errors were compared and analysed in detail. The experiment results show that MQL is inadequate at higher spindle speeds due to its ineffective cooling capacity and weakened lubrication ability. In contrast, CMQL has demonstrated its feasibility and superiority in milling titanium thin-wall parts under all conditions. The outcomes indicate that a lower temperature level of CMQL is advantageous to producing better wear resistance and lower thermomechanical loads, and the CMQL (− 15 ºC) machining environment can remarkably improve the overall machining performance and control the surface form errors of the machined thin-wall parts. At the spindle speed of 3000 rpm, the surface roughness measured under CMQL (− 15 °C) condition is reduced by 16.53% and 23.46%, the deflection value is decreased by 54.74% and 36.99%, while the maximum thickness error is about 53.51% and 20.56% smaller in comparison to flood cooling and MQL machining. In addition, CMQL is an economical and sustainable cooling/lubrication strategy; the outcomes of this work can provide the industry with useful guidance for high-quality machining of thin-wall components.

On-line milling tool wear monitoring under practical machining conditions

Accurate diagnosis of milling tool wear monitoring during high-speed cutting processes is essential for ensuring machining surface precision, enhancing tool efficiency, and extending the service life of machine tools. However, challenges in data screening during each stage of tool movement and signal migration caused by changes in machining parameters have not been sufficiently addressed. To tackle these issues, this paper presents a novel MB-DAAN model, which considers signals from each stage of tool movement and diagnoses tool wear under dynamic machining parameters. Firstly, the multi-branch classification module is introduced, incorporating two independent classifiers to form a multi-branch classification module. This module adaptively extracts features of tool wear states during the cutting stage. Subsequently, the dynamic adversarial factor adjusts the global and local losses between features, aligning conditional and edge distributions across different machining parameters. The models were evaluated in the milling tool wear data set. MB-DAAN achieving a diagnostic accuracy of 97.3%-a significant improvement compared to CNN, DAN, and DAAN models.

Semi-supervised multi-source meta-domain generalization method for tool wear state prediction under varying cutting conditions

Accurate tool wear state prediction during machining is essential for lowering production costs and ensuring quality. Conventional deep learning-based methods perform excellently under constant cutting conditions with sufficient labeled data. During training, domain-adaptation-based transfer learning methods require exposure to partial target domain data to accurately predict under varying cutting conditions. However, the performance of these two methods drops considerably under unseen cutting conditions. Therefore, we propose a semi-supervised multi-source meta-domain generalization (SSM2DG) method for label scarcity training scenarios, which can generalize the deep models to predict tool wear state under unseen cutting conditions. SSM2DG adopts an episodic training strategy based on meta-learning to simulate cutting condition shifts. Then, one basic model based on multiple attention mechanisms is presented for enhancing temporal and cross-channel feature extraction. Next, the entropy-based semi-supervised learning method is incorporated into the episodic training procedure to enrich intra-domain representations using unlabeled data. Besides, class relationship alignment loss based on the distributional sliced-Wasserstein distance and memory centroid-based loss are designed to improve generalization ability. Finally, milling experiments under variable working conditions and multiple ablation studies verify the effectiveness and superiority of the proposed method.

Hierarchical temporal transformer network for tool wear state recognition

The accurate determination of the tool-wear state helps workers maximise tool utilisation while reducing waste. It also ensures the machining quality and improves the machining efficiency. This study proposed a deep learning network based on the self-attention mechanism that enabled global modeling and long-term dependence for better tool wear state recognition by cutting signals. The Hierarchical Temporal Transformer Network (HTT-Net) was constructed by improving the Swin Transformer backbone network to enable the global modeling of the input temporal sequence signal. The token merging layer is used to build a hierarchical feature map, that enables the model to continuously increase the receptive field and improve its global modeling capability. Self-attention calculation is performed on the temporal sequence data partition window, which reduces the complexity of the model from a quadratic into a linear form of the sequence length. The shifted window enables information interaction between non-overlapping windows, which can improve the global modeling capability of the model. PHM2010 public cutting data and TC4 milling data were used for model training and test. The results showed that the tool wear state recognition capability of HTT-Net on the PHM2010 dataset outperformed the models in existing studies, with up to 16.13% improvement in recognition accuracy on the relevant dataset. The recognition accuracy on the TC4 milling dataset can reach 98.87%, which further verified the actual application capability of the model. At last, The Q-Q graph analysis verified that the model had strong robustness and stability, and the ablation experiment verified that each module of the model had positive gain on the recognition results.

Novel extended NI-MWMOTE-based fault diagnosis method for data-limited and noise-imbalanced scenarios

Under real-world conditions, faulty samples of key components (e.g., bearings and cutting tools, etc.) are typically limited and sparse. Additionally, their historical data is characterized by time-series and imbalance characteristics. In other words, the training samples are not only limited and noisy, but also exhibit both within-class and between-class imbalance. These factors present significant challenges in the realm of fault monitoring modeling. To tackle these challenges, this paper presents an innovative fault diagnosis method rooted in the extended NI-MWMOTE and LS-SVM. NI-MWMOTE stands as an advanced noise-immunity majority weighted minority oversampling technique, originally introduced in our prior research, and it has exhibited exceptional competitiveness in noisy imbalanced benchmark datasets. It champions an adaptive noise processing strategy leveraging the distribution characteristics of noisy imbalanced data and the essence of machine learning. Specifically, it employs Euclidean distance and neighbor density to differentiate between spurious noise and true noise, and it determines the optimal processing strategy based on misclassification error and iteration. Furthermore, it employs unsupervised aggregative hierarchical clustering, misclassification error, and majority-weighted minority oversampling in a collaborative manner to address both within-class and between-class imbalanced problems. The primary contribution of our paper lies in the context of the monitoring scenario mentioned above. We have expanded the hyper-parameter range of NI-MWMOTE, corrected and optimized its built-in noise function to enhance the interpretability of the model, and successfully applied it in conjunction with LS-SVM to this particular setting. Notably, this marks the pioneering endeavor within our established knowledge sphere into the domain of tool wear state monitoring. The results suggest that, when compared to 11 well-known algorithms, our framework demonstrates significant competitiveness in real-world scenarios characterized under data-limited and noise-imbalanced scenarios for bearings and cutting tools fault diagnosis. This establishes a solid theoretical and practical foundation for similar scenarios.

Identification of Tool Wear Based on Infographics and a Double-Attention Network

Tool wear is a crucial factor in machining as it directly impacts surface quality and indirectly decreases machining efficiency, which leads to significant economic losses. Hence, monitoring tool wear state is of the utmost importance for achieving high performance and efficient machining. Although monitoring tool wear state using a single sensor has been validated in laboratory settings, it has certain drawbacks such as limited feature information acquisition and inability to learn important features adaptively. These limitations pose challenges to quickly extending the monitoring function of tool wear state of the machine tools. To solve these problems, this paper proposes a double-attention deep learning network based on vibroacoustic signal fusion feature infographics. The first solution is the construction of novel infographics using tool-intrinsic characteristics and multi-domain fusion features of multi-sensor inputs, which includes correlation analysis, principal component analysis, and feature fusion. The second solution is to build a novel deep network with a double-attention module and a spatial pyramid pooling module which can accurately and quickly identify tool wear state by successfully extracting critical spatial data from the infographics at various scales. The validity of the network is examined through an interpretability analysis based on the class activation graph. In terms of the tool wear status recognition task, the F1 score of the double-attention model based on an information graph is 11.61% higher than Resnet18, and peak recognition accuracy reaches 97.98%.

A tool wear condition monitoring method for non-specific sensing signals

Real-time and accurate monitoring of tool wear conditions is crucial to achieving double optimization of production cost and product quality. However, the differences in the characteristics of different signals limit the ability of the monitoring model to generalize between sensing channels, which becomes an important factor limiting the promotion of the model. To solve this problem, an improved parallel residual network based on single-channel and non-specific sensing signals is proposed in this paper. The limitation of the single-channel signal with little information and poor anti-interference ability is overcome by adaptively extracting the multi-scale spatial features of the sensing signal. Hybrid dilated convolution is introduced to expand the receptive field, and then the long historical domain information is obtained. At the same time, the information dependence between layers is enhanced by introducing skip connections. These two designs ensure the perceptual generalization ability of the model. Considering the tool replacement time and the imbalance classification of labels, a comprehensive evaluation method is proposed for model performance evaluation. In addition, the variation law of tool wear in the milling process of Ti-6Al-4V thin-walled parts is investigated. Finally, the validity and transferability of the model are verified by two milling datasets with different cutting conditions. On the basis of ensuring the perceptual generalization ability of the model, the differences in model performance based on acceleration and cutting force signals are controlled within 4.5 % and 1 %, respectively, and the overall average recognition performance is 96.3 % and 92.5 %, respectively. This study provides a feasible solution for intelligent tool replacement in the actual machining environment.

Intelligent Tool Wear Monitoring Method Using a Convolutional Neural Network and an Informer

Tool wear (TW) is the gradual deterioration and loss of cutting edges due to continuous cutting operations in real production scenarios. This wear can affect the quality of the cut, increase production costs, reduce workpiece accuracy, and lead to sudden tool breakage, affecting productivity and safety. Nevertheless, since conventional tool wear monitoring (TWM) approaches often employ complex physical models and empirical rules, their application to complex and non-linear manufacturing processes is challenging. As a result, this study presents a TWM model using a convolutional neural network (CNN), an Informer encoder, and bidirectional long short-term memory (BiLSTM). First, local feature extraction is performed on the input multi-sensor signals using CNN. Then, the Informer encoder deals with long-term time dependencies and captures global time features. Finally, BiLSTM captures the time dependency in the data and outputs the predicted tool wear state through the fully connected layer. The experimental results show that the proposed TWM model achieves a prediction accuracy of 99%. It is able to meet the TWM accuracy requirements of real production needs. Moreover, this method also has good interpretability, which can help to understand the critical tool wear factors.

Tool wear condition monitoring method based on relevance vector machine

Tool wear condition monitoring method based on relevance vector machine

On-line tool wear monitoring under variable milling conditions based on a condition-adaptive hidden semi-Markov model (CAHSMM)

Since tool wear during machining process imposes a significant limitation on production quality as well as efficiency, on-line tool wear monitoring (TWM) is one of considerably crucial issues for intelligent manufacturing system. However, there is still a lack of universal on-line TWM techniques appropriate for the practical production process with variable machining parameters. Therefore, this study proposes a highly condition-adaptive method for tool wear state monitoring and remaining useful life (RUL) estimation under time-varying operation conditions. Multi-domain features are first extracted from vibration, acoustic emission and sound signals, and subsequently fused by the kernel principal component analysis (KPCA) to highlight discriminative information correlated to tool wear. A condition-adaptive hidden semi-Markov model (CAHSMM) is established to describe the evolution of tool wear, in which the accelerated failure time model (AFTM) embedded with Taylor tool life equation is utilized to determine the state transition probabilities that are strongly dependent on cutting parameters and duration time. Moreover, to extrapolate the Gaussian mixture models (GMMs) used to represent the observation probability beyond the range where they were trained, a Moment matching based on-line unsupervised transfer learning (OUTL) method is developed. After obtaining the state transition and observation probabilities, a modified forward algorithm is eventually implemented so as to assess tool wear online. Both the dataset from milling experiments of Ti6Al4V and PHM 2010 are employed for verifying the effectiveness of the proposed method. The results show that the presented method enables to reliably evaluate wear state and RUL of the cutter even under time-switching operation parameters, which demonstrates its promising potential for industrial applications.

Tool Wear Monitoring Based on the Gray Wolf Optimized Variational Mode Decomposition Algorithm and Hilbert–Huang Transformation in Machining Stainless Steel

The online monitoring and prediction of tool wear are important to maintain the stability of machining processes. In most cases, the tool wear condition can be evaluated by signals such as force, sound, vibration, and temperature, which are often processed via Fourier-transform based methods, typically, the short-time Fourier transform (STFT). However, the fixed-width window function in STFT has many limitations. In this paper, a novel tool wear monitoring method based on variational mode decomposition (VMD) and Hilbert–Huang transformation (HHT) were developed to monitor the wear of carbide tools in machining stainless steel. In this method, the intrinsic mode function (IMF) was used as the fitness function, and the (K alpha) parameter sets for VMD were optimized by the gray wolf optimization (GWO). The results show that the characteristic frequency in the GWO-VMD-HHT method is more significant with no aliasing compared with the EMD-HHT method, and an obvious characteristic frequency shift phenomenon is present. By utilizing the energy value of IMF3 as the feature to classify the wear state of the cutting tool, the increase of energy reached 85.48% when 260–315 milling passes were in severe wear state. GWO, which can accurately find the best parameters for VMD, not only solves the problem that the Entropy Function is not suitable for force signals, but also provides reference for the selection of parameters of VMD.

Study on tool wear state monitoring based on EEMD information entropy and PSO-SVM

Tool wear state is closely related to the workpiece surface, so tool wear monitoring has an important engineering significance. This paper proposes a tool for wear monitoring means based on EEMD information entropy and PSO-SVM. Firstly, the force signal of the cutting process collected force by the sensor is decomposed by EEMD, and information entropy of the decomposed signal was used as the tool wear feature quantity. Then the PSO-SVM model was used to classify and identify the extracted tool wear feature quantity, and the recognition accuracy of the method was 97.75%. Compared with BP neural network and SVM model without algorithm optimization, it can quickly and accurately identify three types of tool wear states meanwhile realizing monitoring of tool wear state.