Tool Wear Prediction Research Articles

Research on numerical simulation and prediction of tool wear in cutting ultra-high-strength aluminum alloys

Research on numerical simulation and prediction of tool wear in cutting ultra-high-strength aluminum alloys

Tool wear prediction based on K-means and Adaboost auto-encoder

Abstract A new tool wear prediction model is proposed to address the tool wear issue, aimed at monitoring tool wear based on specific task requirements and guiding tool replacement during actual cutting operations. In the data preprocessing phase, tool wear states are classified using unsupervised K-means clustering. The time, frequency, and time-frequency domain features are then labeled and fused using an autoencoder neural network applied to the original set of signal features from the tool. For tool wear prediction, an enhanced autoencoder neural network leveraging AdaBoost is employed to establish the prediction model. The reconstruction error serves as the chosen loss function to assess the autoencoder's performance, taking into account data correlation and the inherent lossy nature of the autoencoder. Experimental results from real machining data obtained from a CNC milling machine demonstrate that the proposed model achieves higher prediction accuracy while reducing data dimensions.

Bayesian neural networks modeling for tool wear prediction in milling Al 6061 T6 under MQL conditions

Bayesian neural networks modeling for tool wear prediction in milling Al 6061 T6 under MQL conditions

Tool Wear Prediction Combining Global Feature Attention and Long Short-Term Memory Network

This study aims to accurately predict tool flank wear in milling and simplify the complexity of feature selection. A hybrid approach is proposed to eclectically integrate the advantages between the long short-term memory (LSTM) network and the global feature attention (GFA) module. First, the feature matrix is calculated using the multi-domain feature extraction method. Subsequently, a parallel network is employed to achieve feature fusion. The stacked LSTM network extracts the temporal dependencies between features and the GFA module is used to adaptively complement key features representing global information of samples. Finally, the output features are concatenated, and tool wear prediction is achieved through a fully connected layer. Experiments demonstrate the optimal performance in predicting tool flank wear. Specifically, using the proposed GFA-LSTM framework reduces the mean absolute error (MAE) by 36.9%, 17.7%, and 25.2% in three experiments compared to the simple LSTM, validating the effectiveness of the proposed method.

Tool wear prediction in broaching based on tool geometry

Abstract The aircraft engine is a safety-critical part of an aircraft. Constructive modifications resulting in ecological and economic improvements of the engine lead to necessary changes in manufacturing processes. The turbine discs in the low-pressure section are made of temperature resistant nickel-based superalloys. For the connection of the turbine blades with the disc, fir-tree slots are machined by broaching. The broaching process achieves high geometrical accuracy as well as process reliable rim zone properties. Alternative manufacturing processes such as milling or wire electrical discharge machining are still the subject of research in many applications and do not yet achieve the required process reliability. Especially in finishing operations, tool geometry is complex. In previous research work, the local rake angle and, as a result, the acting cutting force were determined analytically and empirically. Based on these results, the relationship between tool geometry, cutting force, and tool wear will be investigated in this work to enable a cutting length dependent prediction of tool wear. The aim of this work is a model-based prediction of tool wear for a given cutting length and a previously unknown tool geometry. For this purpose, a method was developed to predict tool wear using empirical test data as well as internal machine data. With the results of this work, it is not only possible to identify critical points on newly developed broaching tools by maxima of the cutting force, but also to predict wear locally, depending on the cutting length.

Prediction of tool wear and surface finish using ANFIS modelling during turning of Carbon Fiber Reinforced Plastic (CFRP) composites

Carbon fiber-reinforced plastics (CFRP) are widely used in various industries due to their high strength to weight ratio, corrosion resistance, durability, and excellent thermo-mechanical properties. The machining of CFRP composites has always been a challenge for the manufacturers. In this study, CNC turning operation with coated carbide tool is used to machine a specific CFRP and the relationship between the cutting parameters (Speed, Feed, Depth of Cut) and response parameters (Vibration, Surface Finish, Cutting Force and Tool Wear) are investigated. An adaptive-network-based fuzzy inference system (ANFIS) model with two multi-input–single-output (MISO) system has been developed to predict the tool wear and surface finish. Speed, feed, depth of cut, vibration and cutting force have been used as input parameters and tool wear and surface finish have been used as output parameters. Three sets of cutting parameter have been used to gather the data points for continuous turning of CFRP composite. The model merged fuzzy inference modeling with artificial neural network learning abilities, and a set of rules is constructed directly from experimental data. This model is capable of predicting the cutting tool wear and surface finish during turning of CFRP composite. The predicted tool wear and surface finish data are compared to the experimental results. The predicted data agreed well with the actual experimental data with 98.96 % accuracy for tool wear and 99.61 % accuracy for surface finish.

The capability of Acoustic Emission features to monitor diamond-coated burr grinding wear and effectiveness

Within manufacturing there is a growing need for autonomous Tool Condition Monitoring (TCM) systems, with the ability to predict tool wear and failure. This need is increased, when using specialised tools such as Diamond-Coated Burrs (DCBs) for grinding high strength ceramics or glass, in which the random nature of the tool, inconsistent manufacturing methods and high wear rates create large variance in tool life. This unpredictable nature leads to a significant fraction of a DCB tool's life being underutilised due to premature replacement. Workpiece surface damage, increased grinding forces and large-scale diamond grain pullout could all be the result of high levels of runout and in-circularity common within electroplated DCBs. As such it is important to not only monitor the overall tool wear but also tool condition. Acoustic Emission (AE) presents as an indirect on-machine sensing method highly suited to grinding applications. The high frequency range of AE, >20 kHz, prevents machine noise from dominating the acquired signals, isolating the micro-scale machining processes within noises machine environments. AE resulting from the grinding process has the potential to monitor not only tool wear but also runout and circularity. A series of DCB wear tests have been conducted, each consisting of the continuous acquisition of AE during grinding and frequent tool surface measurements. Preliminary results demonstrate AE kurtosis can be seen as an indicator of each tool’s runout, representing the fraction of time the tool and workpiece are in contact during a revolution. As a result, an indirect monitoring system capable of monitoring wear and tool state with AE could be utilised within the manufacturing sector.

Model-based tool wear detection and fault diagnosis for end mill in various cutting conditions

In recent developments in the field of manufacturing systems, there has been a growing emphasis on optimizing cutting conditions. These optimizations are primarily based on intricate parameters, such as the material removal rate (MRR), surface roughness, and position accuracy. Simultaneously, there’s an increasing focus on enhancing manufacturing efficiency through equipment maintenance strategies that consider parameters, such as corrosion, pressure, temperature, vibration, and other environmental factors. Wear is inevitable during processing, which affects productivity. It is generated in various forms, such as flank, crater, and edge wear, which reduce the tool lifespan and impact machining quality, especially by increasing the cutting forces. Various studies have been conducted to address this issue. Direct measurements using microscopes have high accuracy but require interruption during the process, which adversely affects efficiency and productivity. As a solution, the modern era has witnessed an increase in indirect methods. These methods are often sensor-based, capture data during the machining process, and employ various models, including emerging artificial intelligence techniques, for predicting tool wear. However, these methods have problems with environmental susceptibility, reduced reliability, limitations of application, and excessive costs. This paper suggests a tool wear integrated cutting load prediction model, tool wear detection, and fault diagnosis mechanism. The tool-wear-integrated cutting-load prediction model was constructed by combining the cutting-load prediction and tool-wear models. The coefficients of the model were derived from the actual cutting data extracted by the spindle load. Tool wear detection was implemented by dividing regions based on the tendency of the coefficient of the constructed tool wear integrated cutting load prediction model and the errors between the predicted and actual values. The proposed model demonstrated a performance comparable to that of the existing models in a single-cutting-condition path. However, it excelled in extracting the tool wear coefficients in paths with a mixture of various cutting conditions, which was not achievable with conventional models. Based on these coefficients, the cutting force was predicted with a maximum error of 3.3 %. Also, an accurate determination of the tool-wear regions was possible. Furthermore, the performance of the tool fault diagnosis method was validated using images of tools identified as being at risk of damage.

Optimizing tool wear prediction in intelligent manufacturing: a multi-sensor approach enhanced by RealNVP

Optimizing tool wear prediction in intelligent manufacturing: a multi-sensor approach enhanced by RealNVP

Optimization of Tool Wear and Cutting Parameters in SCCO2-MQL Ultrasonic Vibration Milling of SiCp/Al Composites

Silicon carbide particle-reinforced aluminum matrix (SiCp/Al) composites are significant lightweight metal matrix composites extensively utilized in precision instruments and aerospace sectors. Nevertheless, the inclusion of rigid SiC particles exacerbates tool wear in mechanical machining, resulting in a decline in the quality of surface finishes. This work undertakes a comprehensive investigation into the problem of tool wear in SiCp/Al composite materials throughout the machining process. Initially, a comprehensive investigation was conducted to analyze the effects of cutting velocity vc, feed per tooth fz, milling depth ap, and milling width ae on tool wear during high-speed milling under SCCO2-MQL (Supercritical Carbon Dioxide Minimum Quantity Lubrication) ultrasonic vibration conditions. The results show that under the condition of SCCO2-MQL ultrasonic vibration, proper control of milling parameters can significantly reduce tool wear, extend tool service life, improve machining quality, and effectively reduce blade breakage and spalling damage to the tool, reduce abrasive wear and adhesive wear, and thus significantly improve the durability of the tool. Furthermore, a prediction model for tool wear was developed by employing the orthogonal test method and multiple linear regression. The model’s relevance and accuracy were confirmed using F-tests and t-tests. The results show that the model can effectively predict tool wear, among which cutting velocity vc and feed rate fz are the key parameters affecting the prediction accuracy. Finally, a genetic algorithm was used to optimize the milling parameters, and the optimal parameter combination (vc = 60.00 m/min, fz = 0.08 mm/z, ap = 0.20 mm) was determined, and the optimized milling parameters were tested. Empirical findings suggest that the careful selection of milling parameters can significantly mitigate tool wear, extend the lifespan of the tool, and enhance the quality of the surface. This work serves as a significant point of reference for the processing of SiCp/Al composite materials.

Enhancement of tool life using magneto-rheological fluid damping and tool wear prediction through deep learning model in milling

Undesirable vibrations generated during milling significantly intensify milling tool wear and reduces tool life. Hence it is imperative to employ the effective vibration control strategy during milling to optimize the overall effectiveness of the process. Thus, the present study incorporates the magneto-rheological fluid damping for vibration mitigation during the milling to enhance the life of a cutting tool under different operating conditions. Simultaneously, the data-driven modelling are employed to predict the tool wear using real-time multi-sensor machining data under different operating conditions. Accordingly, the real-time multi-sensor data was acquired at different operating conditions during normal and vibration-damped milling. Later, the extracted tine-domain features were selected using the sequential feature selection with Random Forest Regressor to obtain the most relevant time-domain features from multi-sensor data for improving the model performance. These selected features are then fed to the different deep-learning algorithms for tool wear prediction. Thus, Magneto-rheological fluid damping significantly reduces the vibrations and cutting forces in vibration-damped milling between 65 and 80% and improves the tool life by 18–21 % compared to normal milling under different operating conditions. Moreover, among the different deep learning algorithms, hybrid Encoder-Decoder Long short-term Memory is considered the optimal choice for tool wear prediction due to highest accuracy with coefficient of determinations (R2) ranging from 0.980 to 0.995 with the lowest mean squared error between 22 and 50 μm under varying operating conditions during normal and vibration-damped milling. Thus, it facilitates the significant enhancement in the tool life under different operating conditions, and robust data-driven model for the effective implementation of predictive maintenance approach.

Intelligent tool wear prediction based on deep learning PSD-CVT model

To ensure the reliability of machining quality, it is crucial to predict tool wear accurately. In this paper, a novel deep learning-based model is proposed, which synthesizes the advantages of power spectral density (PSD), convolutional neural networks (CNN), and vision transformer model (ViT), namely PSD-CVT. PSD maps can provide a comprehensive understanding of the spectral characteristics of the signals. It makes the spectral characteristics more obvious and makes it easy to analyze and compare different signals. CNN focuses on local feature extraction, which can capture local information such as the texture, edge, and shape of the image, while the attention mechanism in ViT can effectively capture the global structure and long-range dependencies present in the image. Two fully connected layers with a ReLU function are used to obtain the predicted tool wear values. The experimental results on the PHM 2010 dataset demonstrate that the proposed model has higher prediction accuracy than the CNN model or ViT model alone, as well as outperforms several existing methods in accurately predicting tool wear. The proposed prediction method can also be applied to predict tool wear in other machining fields.

Utilizing TGAN and ConSinGAN for Improved Tool Wear Prediction: A Comparative Study with ED-LSTM, GRU, and CNN Models

Developing precise deep learning (DL) models for predicting tool wear is challenging, particularly due to the scarcity of experimental data. To address this issue, this paper introduces an innovative approach that leverages the capabilities of tabular generative adversarial networks (TGAN) and conditional single image GAN (ConSinGAN). These models are employed to generate synthetic data, thereby enriching the dataset and enhancing the robustness of the predictive models. The efficacy of this methodology was rigorously evaluated using publicly available milling datasets. The pre-processing of acoustic emission data involved the application of the Walsh-Hadamard transform, followed by the generation of spectrograms. These spectrograms were then used to extract statistical attributes, forming a comprehensive feature vector for model input. Three DL models—encoder-decoder long short-term memory (ED-LSTM), gated recurrent unit (GRU), and convolutional neural network (CNN)—were applied to assess their tool wear prediction capabilities. The application of 10-fold cross-validation across these models yielded exceptionally low RMSE and MAE values of 0.02 and 0.16, respectively, underscoring the effectiveness of this approach. The results not only highlight the potential of TGAN and ConSinGAN in mitigating data scarcity but also demonstrate significant improvements in the accuracy of tool wear predictions, paving the way for more reliable and precise predictive maintenance in manufacturing processes.

TOOL CONDITION MONITORING IN MACHINING USING ROBUST RESIDUAL NEURAL NETWORKS

Tool condition monitoring (TCM) aims to improve process efficiency, quality and tool maintenance costs by monitoring critical variables such as tool wear. This study proposes a deep learning (DL) architecture based on process-informed robust residual networks (Robust-ResNet) to predict tool wear in milling processes using time series of internal computer numerical control (CNC) signals. The Robust-ResNet architecture uses skip connections to move through multiple convolutional layers, avoiding the vanishing gradient problem of other neural network algorithms. The study includes an evaluation of the binding of process information as input to the architecture and an attention mechanism between skips to make more robust predictions. The proposed architecture has been trained and optimised using an open access data set of face milling time series. In this particular case, AC and DC signals have been used together with the corresponding tool wear values. The results of this study demonstrate the benefits of using deep learning techniques in the prediction of tool wear using internal signals provided by the CNC itself. The implementation of the proposed architecture is expected to help reduce maintenance costs, improve product quality and increase production efficiency in milling manufacturing processes.

Milling tool wear prediction using an integrated wireless multi-sensor tool holder and convolutional neural networks

Milling tool wear prediction using an integrated wireless multi-sensor tool holder and convolutional neural networks

Gated recurrent unit and temporal convolutional network with soft thresholding and attention mechanism for tool wear prediction

Predicting tool wear significantly improves cutting efficiency and quality in intelligent manufacturing. Traditional recurrent neural networks suffer from vanishing or exploding gradients and need to be improved prediction accuracy. Therefore, this paper proposes a gated recurrent unit and temporal convolutional network with soft thresholding and attention mechanism (TCN-BiGRU-SA). Monotonicity, robustness, and trendability indicators are used to assess time and frequency domain features. Then, evaluation results of these three indicators are weighted to identify the sensitive features with the degradation trend. The attention mechanism is proposed to adaptively adjust the soft thresholding, and the SA is added to TCN-BiGRU to remain helpful features. Finally, the proposed model is validated using an experimental dataset. Compared to the TCN-BiGRU model without the SA mechanism, the predicted results show that the average mean absolute error and root mean square error are reduced by 48.29 % and 36.39 %, respectively.

Tool wear prediction method based on dual-attention mechanism network

ABSTRACT Effective tool wear monitoring (TWM) methods are helpful in accurately estimating tool wear status, and rationally using and replacing tools, thereby improving machining quality and production efficiency. In this paper, a dual-attention mechanism network is designed, which not only overcomes the information loss problem, but also enhances the interpretability of deep learning model. Firstly, Z-Score standardization is applied to the cutting force signals, CNN-GRU is built as the basic framework, improving the traditional SE attention mechanism, and integrating it into the model input for adaptive weight allocation. Furthermore, embedding the multi-head attention mechanism into the connection layer between CNN and GRU, and the weighted features are input into GRU for sequence modeling. Ultimately, a fully connected layer is used to establish a mapping from high-dimensional features to the dimension of tool wear, and a linear regression layer is used to output wear prediction values. Compared with several common deep learning models, the dual-attention mechanism model is more reliable and superior. The enhancement of model performance through the dual-attention mechanism was further quantified through the ablation experiment, specifically, the MAE was reduced by 62.33%, 46.03%, and 33.58%, and the RMSE was reduced by 63.05%, 48.62%, and 35.70%, respectively.

A meta-learning method for smart manufacturing: Tool wear prediction using hybrid information under various operating conditions

Accurate tool wear prediction during machining is crucial to manufacturing since it will significantly influence tool life, machining efficiency, and workpiece quality. Although existing data-driven methods can achieve competitive performance in tool wear prediction, their main emphasis is on fixed operating conditions with sufficient training samples, which is impractical in engineering practice. This implies that predicting tool wear values under variable working conditions with insufficient data is still a challenge owing to the difference in data distributions in complex tool wear mechanisms. Besides, having no access to samples in new conditions is another challenge for tool wear prediction in engineering practice. To address these issues, we develop a hybrid information model-agnostic domain generalization (H-MADG) method to provide appropriate initial model parameters that can be fast adaptative to the new conditions after fine-tuning. Additionally, we construct hybrid information as model input by fusing process information with temporal properties derived by neural networks, and the hybrid information can offer more useful prior knowledge about the machining process. Experimental results on NASA milling data show that compared with contrastive techniques, the RMSE of the proposed H-MADG method is reduced by an average of 36.81 %, which can achieve a low average RMSE value of 0.0904 with 15 cases under five different network architectures. We also investigate several crucial impact factors of the H-MADG method and summarize corresponding analysis and suggestions.

Physics-Informed Neural Networks With Weighted Losses by Uncertainty Evaluation for Accurate and Stable Prediction of Manufacturing Systems.

The state prediction of key components in manufacturing systems tends to be risk-sensitive tasks, where prediction accuracy and stability are the two key indicators. The physics-informed neural networks (PINNs), which integrate the advantages of both data-driven models and physics models, are deemed as an effective approach and research trends for stable prediction; however, the potential advantages of PINN are limited for the situations with inaccurate physics models or noisy data, where the balancing of the weights of the data-driven model and physics model is very important for improving the performance of PINN, and it is also a challenge urgently to be addressed. This article proposed a kind of PINN with weighted losses (PNNN-WLs) by uncertainty evaluation for accurate and stable prediction of manufacturing systems, where a novel weight allocation strategy based on uncertainty evaluation by quantifying the variance of prediction errors is proposed, and an improved PINN framework is established for accurate and stable prediction. The proposed approach is verified with open datasets on tool wear prediction, and experimental results show that the prediction accuracy and stability could be obviously improved over existing methods.

Machine Learning based Tool Wear Prediction from Variability of Acoustic Sound Emission Signals

Machine Learning based Tool Wear Prediction from Variability of Acoustic Sound Emission Signals