Tool Wear Estimation Research Articles

실시간 절삭 계수 식별 기반의 공구 마모 모니터링 시스템

Among the monitoring technologies in the metal-cutting process, tool wear is the most critical monitoring factor in real machining sites. Extensive studies have been conducted to monitor equipment breakdown in real-time. For example, tool wear prediction studies using cutting force signals and deducting force coefficient values from the cutting process. However, due to many limitations, those wearable monitoring technologies have not been directly adopted in the field. This paper proposes a novel tool wear predictor using the cutting force coefficient with various cutting tools, and its validity evaluates through cutting tests. Tool wear prediction from the cutting force coefficient should conduct in real-time for adoption in real machining sites. Therefore, a real-time calculation algorithm of the cutting force coefficient and a tool wear estimation method proposes, and they compare with actual tool wear in cutting experiments for validation. Validation cutting tests are conducted with carbon steel and titanium, the most commonly used materials in real cutting sites. In future work, validation will be conducted with different materials and cutting tools, considering the application in real machining sites.

A transformer‐based end‐to‐end data‐driven model for multisensor time series monitoring of machine tool condition

AbstractOnline determination of a cutter's health status is crucial for the attainment of condition‐based automated tool change in computer numerically controlled (CNC) machining. Due to the impracticalities associated with direct condition measurements, data‐based modeling of monitoring signals provides a viable practical route. However, the highly noisy and redundant nature of the associated data impacts negatively on model's accuracy and typically calls for additional initial preprocessing before modeling. Additionally, the long sequential data entails widely varying condition distributions exhibited by different cutters, even from the same batch on similar machining parameters, posing a challenge to model generalization. An end‐to‐end model has thus been developed to work directly on unprocessed data to establish global sensitive features from varying distributions for online tool wear estimation in CNC machining. The model utilizes three main functional blocks. First, a data denoising and feature selection block automatically processes raw multisensor data directly, dispensing with scaling or preprocessing of inputs as conventionally done. Each sensor channel's independence is preserved at initial processing ensuring complementary information from different sensors is utilized while simultaneously minimizing existing redundancies. The weighted denoised data is then processed through a transformer encoder block for determination of global dependencies in the time‐series sequence, regardless of the time‐step position. The learned features are then fed to an upper supervised learning block for association with the monitored wear condition. The developed model works directly on raw noisy data irrespective of scaling differences, saving on preprocessing computational cost. The global associations extracted on long sequences by the transformer‐encoder allow for model generalization to varying wear distributions. The parallel processing structure of all channels ensures complementary information is utilized minimizing unforeseen model bias. The model's performance as evaluated on experimental milling data and further comparison with other reported models on same dataset shows attainment of comparable state‐of‐art results.

Online tool wear monitoring by super-resolution based machine vision

The tool condition has been a major concern in modern computer numerical control (CNC) machining due to its direct effects on the quality of final product both in the surface and dimensional integrity. The conventional machine vision-based tool condition monitoring (TCM) approaches cannot meet the high precision requirement in micro machining, as the cutting parameters are in micro scale and the spindle works in high rotation speed which makes online tool wear measurement quite difficult. To meet these challenges, this study develops a single image super-resolution (SISR) approach for direct tool wear estimation in micro-milling. Motivated by the self-similarity of tool wear image morphology, this study proposes a sparse decomposition framework by learning dictionaries from the tool wear image pyramid. Based on their multi-scale invariant properties, the similar image patches of coarse scales can be retrieved from fine scales to reconstruct the high-resolution image fastly and in high quality. The reconstructed high-resolution image then can be conveniently applied to wear monitoring, and overcomes the image acquisition deficiencies of the conventional machine vision-based monitoring approaches. Experimental results validate this approach for the tool wear area estimation as well as its generalization of the wear width with regarding to the conventional manual measurements. • Develop a single image super-resolution approach for direct tool wear estimation. • Propose a sparse decomposition framework by learning multiscale dictionaries of the tool wear image. • Achieve image super-resolution reconstruction fastly based on multi-scale invariant feature retrieval. • The reconstructed SR image can be conveniently applied to wear monitoring.

Tool wear estimation with a data-driven physics coupling approach

It’s essential to predict tool wear condition in high-speed milling (HSM), which is a key factor affecting machining precision and part quality. Presently, many physics-based methods and data-driven machine learning methods are applied for tool wear prediction. Inspired by the previously developed generic physical model, a novel data-driven physics coupling approach is established and applied to identify and predict tool flank wear in this study. The proposed ARIMA-WNN model is verified with case studies, in which the determination coefficients are stable at above 95%. It’s flexible to choose the position and length of leading time intervals and predict tool wear condition, which is more adaptive and effective than physics-based methods. In addition, it clarifies the specific distribution of each parameter to machining conditions in the generic physical-empirical wear model.

A Novel Multi-Task Learning Model with PSAE Network for Simultaneous Estimation of Surface Quality and Tool Wear in Milling of Nickel-Based Superalloy Haynes 230.

For data-driven intelligent manufacturing, many important in-process parameters should be estimated simultaneously to control the machining precision of the parts. However, as two of the most important in-process parameters, there is a lack of multi-task learning (MTL) model for simultaneous estimation of surface roughness and tool wear. To address the problem, a new MTL model with shared layers and two task-specific layers was proposed. A novel parallel-stacked auto-encoder (PSAE) network based on stacked denoising auto-encoder (SDAE) and stacked contractive auto-encoder (SCAE) was designed as the shared layers to learn deep features from cutting force signals. To enhance the performance of the MTL model, the scaled exponential linear unit (SELU) was introduced as the activation function of SDAE. Moreover, a dynamic weight averaging (DWA) strategy was implemented to dynamically adjust the learning rate of different tasks. Then, the time-domain features were extracted from raw cutting signals and low-frequency reconstructed wavelet packet coefficients. Frequency-domain features were extracted from the power spectrum obtained by the Fourier transform. After that, all features were combined as the input vectors of the proposed MTL model. Finally, surface roughness and tool wear were simultaneously predicted by the trained MTL model. To verify the superiority and effectiveness of the proposed MTL model, nickel-based superalloy Haynes 230 was machined under different cutting parameter combinations and tool wear levels. Some other intelligent algorithms were also implemented to predict surface roughness and tool wear. The results showed that compared with the support vector regression (SVR), kernel extreme learning machine (KELM), MTL with SDAE (MTL_SDAE), MTL with SCAE (MTL_SCAE), and single-task learning with PSAE (STL_PSAE), the estimation accuracy of surface roughness was improved by 30.82%, 16.67%, 14.06%, 26.17%, and 16.67%, respectively. Meanwhile, the prediction accuracy of tool wear was improved by 46.74%, 39.57%, 41.51%, 38.68%, and 39.57%, respectively. For practical engineering application, the dimensional deviation and surface quality of the machined parts can be controlled through the established MTL model.

Three-Stage Wiener-Process-Based Model for Remaining Useful Life Prediction of a Cutting Tool in High-Speed Milling.

Tool condition monitoring can be employed to ensure safe and full utilization of the cutting tool. Hence, remaining useful life (RUL) prediction of a cutting tool is an important issue for an effective high-speed milling process-monitoring system. However, it is difficult to establish a mechanism model for the life decreasing process owing to the different wear rates in various stages of cutting tool. This study proposes a three-stage Wiener-process-based degradation model for the cutting tool wear estimation and remaining useful life prediction. Tool wear stages classification and RUL prediction are jointly addressed in this work in order to take full advantage of Wiener process, as this three-stage Wiener process definitely constitutes to describe the degradation processes at different wear stages, based on which the overall useful life can be accurately obtained. The numerical results obtained using extensive experiment indicate that the proposed model can effectively predict the cutting tool’s remaining useful life. Empirical comparisons show that the proposed model performs better than existing models in predicting the cutting tool RUL.

Intelligent Health Monitoring of Machine Tools Using a Bayesian Multibranch Neural Network

Online health monitoring of machine tools is an essential technique for tool life extension, manufacturing productivity improvement and product quality improvement. In the era of industrial big data, numerous deep learning (DL)-based methods have been proposed to achieve these goals. However, in complex and dynamic manufacturing processes, practical concerns such as uncertainty quantification and anti-noise capabilities of the DL-based methods are rarely considered. Thus, in this study, a novel multi-branch Bayesian Neural Network (BNN) is developed for the reliable and robust online health monitoring of Computer Numerical Control (CNC) machine tools. With the proposed model, the heterogeneous fault information extracted from multiple sensors can be simultaneously integrated in a deep convolutional neural network (DCNN)-multiple layer perceptron (MLP)-based multi-branch neural network to enhance the health monitoring accuracy and robustness. Furthermore, the proposed multi-branch neural network is extended into a BNN to improve its uncertainty quantification capabilities. The proposed method is evaluated on the tool wear tests of three cutting tools. Tool wear estimation results indicate that the proposed method outperforms comparative methods and achieves the best prediction accuracy and robustness on all three health monitoring tasks investigated in this study. We also found that the proposed method can accurately classify tool wear stages and reach up to 95% mean classification accuracy, which is the best among comparative methods. Also, measures, such as coverage probability of estimation interval (EICP) and normalized mean estimation interval width (NMEIW), are used to assess the capability of quantifying the confidence intervals (CIs) of the tool wear estimations. Results show that the proposed method achieves superior CIs quantification performance with the average EICP and NMEIW values of 95.77% and 0.27 on all three health monitoring tasks.

Tool wear estimation using a CNN-transformer model with semi-supervised learning

In the machining industry, tool wear has a great influence on machining efficiency, product quality, and production costs. To achieve accurate tool wear estimation, a novel CNN-transformer neural network (CTNN) model is proposed in this paper. In the CTNN model, the transformer model and convolutional neural networks (CNN) are used to process condition monitoring (CM) data in parallel, such as cutting force. The motivations are as follows. For one thing, both the transformer model and CNN can extract useful temporal features from CM data, and the learned temporal features by these two parts are fused to achieve accurate tool wear estimation. For another, CNN contributes to enhancing the transformer’s ability to capture the sequence order. In addition, data noise introduces the aleatoric uncertainty to the estimation results. To quantify the aleatoric uncertainty, a negative log-likelihood loss function is employed to enable the model to output the probabilistic distribution associated with tool wear. In such cases, the model outputs both the tool wear and variance, and the variance is learned within the model in an unsupervised manner. Finally, the effectiveness and superiority of the proposed method are validated on a public milling dataset. It is found by experiments that both the transformer model and CNN play important roles in tool wear estimation, and better performance can be obtained when they are used in parallel. In summary, the experimental results suggest that the proposed model can obtain promising results in tool wear estimation.

A Novel Machine Learning-Based Methodology for Tool Wear Prediction Using Acoustic Emission Signals.

There is an increasing trend in the industry of knowing in real-time the condition of their assets. In particular, tool wear is a critical aspect, which requires real-time monitoring to reduce costs and scrap in machining processes. Traditionally, for the purpose of predicting tool wear conditions in machining, mathematical models have been developed to extract the information from the signal of sensors attached to the machines. To reduce the complexity of developing physical models, where an in-depth knowledge of the system being modelled is required, the current trend is to use machine-learning (ML) models based on data from the tool wear. The acoustic emission (AE) technique has been widely used to capture data from and understand the real-time condition of industrial assets such as cutting tools. However, AE signal interpretation and processing is rather complex. One of the most common features extracted from AE signals to predict the tool wear is the counts parameter, defined as the number of times that the amplitude of the signal exceeds a predefined threshold. A recurrent problem of this feature is to define the adequate threshold to obtain consistent wear prediction. Additionally, AE signal bandwidth is rather wide, and the selection of the optimum frequencies band for feature extraction has been pointed out as critical and complex by many authors. To overcome these problems, this paper proposes a methodology that applies multi-threshold count feature extraction at multiresolution level using wavelet packet transform, which extracts a redundant and non-optimal feature map from the AE signal. Next, recursive feature elimination is performed to reduce and optimize the vast number of predicting features generated in the previous step, and random forests regression provides the estimated tool wear. The methodology presented was tested using data captured when turning 19NiMoCr6 steel under pre-established cutting conditions. The results obtained were compared with several ML algorithms such as k-nearest neighbors, support vector machines, artificial neural networks and decision trees. Experimental results show that the proposed method can reduce the predicted root mean squared error by 36.53%.

Tool wear condition monitoring in milling process based on data fusion enhanced long short-term memory network under different cutting conditions

Tool wear condition monitoring (TCM) is essential for milling process to ensure the machining quality, and the long short-term memory network (LSTM) is a good choice for predicting tool wear value. However, the robustness of LSTM- based method is poor when cutting condition changes. A novel method based on data fusion enhanced LSTM is proposed to estimate tool wear value under different cutting conditions. Firstly, vibration time series signal collected from milling process are transformed to feature space through empirical mode decomposition, variational mode decomposition and fourier synchro squeezed transform. And then few feature series are selected by neighborhood component analysis to reduce dimension of the signal features. Finally, these selected feature series are input to train the bidirectional LSTM network and estimate tool wear value. Applications of the proposed method to milling TCM experiments demonstrate it outperforms significantly SVR- based and RNN- based methods under different cutting conditions.

Estimation of Tool Wear and Surface Roughness Development Using Deep Learning and Sensors Fusion.

This paper proposes an estimation approach for tool wear and surface roughness using deep learning and sensor fusion. The one-dimensional convolutional neural network (1D-CNN) is utilized as the estimation model with X- and Y-coordinate vibration signals and sound signal fusion using sensor influence analysis. First, machining experiments with computer numerical control (CNC) parameters are designed using a uniform experimental design (UED) method to guarantee the variety of collected data. The vibration, sound, and spindle current signals are collected and labeled according to the machining parameters. To speed up the degree of tool wear, an accelerated experiment is designed, and the corresponding tool wear and surface roughness are measured. An influential sensor selection analysis is proposed to preserve the estimation accuracy and to minimize the number of sensors. After sensor selection analysis, the sensor signals with better estimation capability are selected and combined using the sensor fusion method. The proposed estimation system combined with sensor selection analysis performs well in terms of accuracy and computational effort. Finally, the proposed approach is applied for on-line monitoring of tool wear with an alarm, which demonstrates the effectiveness of our approach.

New Tool Wear Estimation Method of the Milling Process Based on Multisensor Blind Source Separation

Timely and effective identification and monitoring of tool wear is important for the milling process. However, traditional methods of tool wear estimation have run into difficulties due to under small samples with less prior knowledge. This article addresses this issue by employing a multisensor tool wear estimation method based on blind source separation technology. Stationary subspace analysis (SSA) technology is applied to transform multisensor signals to stationary and nonstationary sources without prior information of signals. Ten dimensionless time-frequency indices of the nonstationary signal are extracted to train least squares support vector regression (LS-SVR) to obtain a tool wear estimation model for small samples. The analysis and comparison of one benchmark tool wear dataset and tool wear experiments verify the feasibility and effectiveness of the proposed method and outperform other two current methods.

In-process stochastic tool wear identification and its application to the improved cutting force modeling of micro milling

• An improved integrated PF-LSTM identification methodology is proposed to predict stochastic tool wear values of micro milling. • The proposed identification method bridges the gap between the in-process stochastic tool wear progression and historical measured data. • On account of the acquired in-process tool wear, the cutting force model is modified in ploughing- and shearing-dominant regimes. • The process nonlinearities are combined with the in-process stochastic tool wear values to estimate the instantaneous uncut chip thickness. Micro milling aims to manufacture miniature structures with high quality and complex features, and the stochastic time-varying tool wear is a crucial factor which has great influence on machining quality and efficiency of micro milling process. To improve the precision of machining and sustainability of micro cutting tools, the in-process tool wear conditions should be identified and updated ahead of time. In this work, an improved integrated estimation method is proposed based on the long short-term memory (LSTM) network and particle filter (PF) algorithm to predict the stochastic tool wear values. The integrated PF-LSTM identification methodology is developed to predict the in-process stochastic tool wear progression on the basis of the historical measurement data. With the estimation of in-process stochastic tool wear, the cutting force model is modified, in which the influence of tool run-out and the trochoidal trajectory of cutting edge are also considered. The proposed integrated estimation method of in-process stochastic tool wear and the modified cutting force model were validated by the micro milling experiments with workpiece material Al6061. It can be seen from the comparison results that the availability and sustainability of micro cutting tool have been improved, and the prediction accuracy also could be increased by 3.4% compared with that without considering the influence of tool wear.

Estimation and Optimization of Tool Wear in Conventional Turning of 709M40 Alloy Steel Using Support Vector Machine (SVM) with Bayesian Optimization.

Cutting tool wear reduces the quality of the product in production processes. The optimization of both the machining parameters and tool life reliability is an increasing research trend to save manufacturing resources. In the present work, we introduced a computational approach in estimating the tool wear in the turning process using artificial intelligence. Support vector machines (SVM) for regression with Bayesian optimization is used to determine the tool wear based on various machining parameters. A coated insert carbide tool 2025 was utilized in turning tests of 709M40 alloy steel. Experimental data were collected for three machining parameters like feed rate, depth of cut, and cutting speed, while the parameter of tool wear was calculated with a scanning electron microscope (SEM). The SVM model was trained on 162 experimental data points and the trained model was then used to estimate the experimental testing data points to determine the model performance. The proposed SVM model with Bayesian optimization achieved a superior accuracy in estimation of the tool wear with a mean absolute percentage error (MAPE) of 6.13% and root mean square error (RMSE) of 2.29%. The results suggest the feasibility of adopting artificial intelligence methods in estimating the machining parameters to reduce the time and costs of manufacturing processes and contribute toward greater sustainability.

Process data based estimation of tool wear on punching machines using TCN-Autoencoder from raw time-series information

Tracking the wear states of tools on punching machines is necessary to reduce scrap rates. In this paper, we propose a method to estimate wear state of punches using Temporal Convolutional Network Autoencoder (TCN-Autoencoder), one of the deep learning techniques for learning time-series information with convolutional architecture. Approach involves inputting raw time-series information, such as sensor, vibration and audio data, into TCN-Autoencoder, and calculating the reconstruction error between the output and the input data. The reconstruction error is used as “anomaly score” and indicates the distance from the normal state. By training TCN-Autoencoder only with data annotated as “normal” state, the reconstruction error becomes larger when inputting abnormal state data, which corresponds the wear state of the punch. Performance is evaluated on experimental measurement data that spans various wear states of the punch. The results showed our model can estimate anomalies faster than the conventional machine-learning-based anomaly estimation method, while maintaining the high estimation accuracy. This is due to TCN-Autoencoder being able to learn from both frequency and time domain.

Vibration-based tool condition monitoring in milling of Ti-6Al-4V using an optimization model of GM(1,N) and SVM

Titanium alloys are the difficult to cut metals due to their low thermal conductivity and chemical affinity with tool material. Since the tool vibration is a replica of tool wear and surface roughness, the present study has proposed a methodology for estimating tool wear and surface roughness based on tool vibration for milling of Ti-6Al-4V alloy using cemented carbide mill cutter. Experiments are conducted at optimum levels of cutting speed, feed per tooth, and depth of cut, and experimental results for the tool vibration, tool wear, and surface roughness are collected until the flank wear reached 0.3 mm (ISO3685:1993). In the next stage, an optimization model of grey prediction GM(1,N) system and support vector machine (SVM) are used and estimated tool wear and surface roughness related to tool vibration. The predicted values of tool wear and surface roughness are compared with the experimental results. The optimization model of GM(1,N) predicted the tool wear and surface roughness with an average error of 3.03% and as 0.7% respectively while the SVM predicted with an average error of 7.67% and 4.45%, respectively.

In-situ tool wear area evaluation in micro milling with considering the influence of cutting force

In order to realize the tool wear in-situ monitoring in micro milling, a novel two-dimensional tool wear estimation approach is developed in this work. The novelty and strong point of the approach is that it can achieve both high estimation accuracy and computational efficiency for fast tool condition monitoring. For this purpose, an empirical statistical model including both process parameters and force features is firstly proposed for in-situ tool wear area estimation. Then the model is improved to enhance its practicability. By comparing the experimental measurements against the results predicted by the improved model and neural network model, it is shown that the improved model has better prediction effect, which illustrates that this approach can realize tool wear estimation in micro milling. Finally, the influence of each variable in improved model on tool wear is analyzed by grey relational degree. The results of this study indicate that this approach can be used to optimize cutting parameters and predict tool wear online.

A GAPSO-Enhanced Extreme Learning Machine Method for Tool Wear Estimation in Milling Processes Based on Vibration Signals

Tools are the most vulnerable components in milling processes conducted using numerical control milling machines, and their wear condition directly influences work-product quality and operational safety. As such, tool wear estimation is an essential component of NC milling operations. This study addresses this issue by proposing an extreme learning machine (ELM) method enhanced by a hybrid genetic algorithm and particle swarm optimization (GAPSO) approach for conducting tool wear estimation based on workpiece vibration signals. Here, a few feature parameters in the time, frequency, and time–frequency (Ensemble empirical mode decomposition, EEMD) domains of the workpiece vibration signals are extracted as the input of the ELM model. Then, the initialized weights and thresholds of the ELM model are optimized based on the GAPSO approach with training dataset. Finally, tool wear is estimated using the optimized ELM model with testing dataset. The effectiveness of the proposed method is verified by its application to vibration signals collected from two milling tool wear experiments (an open-access benchmark dataset and a milling tool wear experiment) by comparison to the ELM, GA-ELM, and PSO-ELM methods. The results indicate that the estimation accuracy and optimization efficiency of the proposed method outperforms that of other three methods.

Tool wear estimation and life prognostics in milling: Model extension and generalization

Tool wear condition is a key factor in milling which directly affects machining precision and part quality. It is essential to seek a convenient method to model and predict tool states. A generic wear model with adjustable coefficients is proposed and validated in this study. Considering the inner mechanisms of different wear stages, the entire tool life is split into three mainly wear zones by critical time, which correspond to three main types of wear: running-in wear, adhesive wear, and three-body abrasive wear. The wear model is validated based on the experimental data, compared with other celebrated wear models, and then further improved to enhance the adaptability and generalization. It is shown that the generalized wear model can discriminate tool wear ranges accurately. The determination coefficient of the wear model is more than 98% with the experimental data. Based on the proposed model, an approach for tool life prognosing and tool wear condition evaluating is proposed. The predictive real-time monitoring data of tool life and wear can be obtained timely with a genetic algorithm.

Data-Driven Remaining Useful Life Estimation for Milling Process: Sensors, Algorithms, Datasets, and Future Directions

An increase in unplanned downtime of machines disrupts and degrades the industrial business, which results in substantial credibility damage and monetary loss. The cutting tool is a critical asset of the milling machine; the failure of the cutting tool causes a loss in industrial productivity due to unplanned downtime. In such cases, a proper predictive maintenance strategy by real-time health monitoring of cutting tools becomes essential. Accurately predicting the useful life of equipment plays a vital role in the predictive maintenance arena of industry 4.0. Many active research efforts have been done to estimate tool life in varied directions. However, the consolidated study of the implemented techniques and future pathways is still missing. So, the purpose of this paper is to provide a systematic and comprehensive literature survey on the data-driven approach of Remaining Useful Life (RUL) estimation of cutting tools during the milling process. The authors have summarized different monitoring techniques, feature extraction methods, decision-making models, and available sensors currently used in the data-driven model. The authors have also presented publicly available datasets related to milling under various operating conditions to compare the accuracy of the prediction model for tool wear estimation. Finally, the article concluded with the challenges, limitations, recent advancements in RUL prognostics techniques using Artificial Intelligence (AI), and future research scope to explore more in this area.