Wear Prediction Research Articles

Tool wear prediction method based on bidirectional long short-term memory neural network of single crystal silicon micro-grinding

Tool wear prediction method based on bidirectional long short-term memory neural network of single crystal silicon micro-grinding

Joint multi-objective dynamic scheduling of machine tools and vehicles in a workshop based on digital twin

Dynamic job-shop scheduling is vital for improving job-shop efficiency and reducing energy consumption. A workshop is a complex combination of multiple functions. Previous studies regarding workshop scheduling showed several problems, such as delayed rescheduling response, single influencing factors, and the separation of machine tools and vehicles. Therefore, to address these issues, this paper constructs a multi-factor scheduling service system with various factors, such as machine failure, tool wear, and product quality. Combined with the international background of rising energy prices, the scheduling scheme is evaluated while having two goals to achieve: minimizing the completion time and achieving energy consumption flexibly by controlling the different parameters’ values. Introducing digital twin theory, machine fault prediction, tool wear prediction, and product quality monitoring can achieve the timeliness and predictability of the workshop scheduling, the joint scheduling of machine tools, and the vehicle transport tasks. Finally, the superiority of this method is verified by processing some critical parts in a marine diesel engine workshop.

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.

Analysis of the causes of severe side wear of the high rail on metro curves by numerical simulation and field investigation

Two of the Shanghai metro lines suffered from severe side wear on the high rail, not only in sharp curves but also in large curves, and the service life of some rails on sharp curves was reduced by approximately 50% compared with the last rail replacement. Features and evolution rules for the rail side wear and the wheel flange wear were determined by field measurements and long-term monitoring. A rail wear prediction model, verified in our previous work, is used to reveal the influence of factors on rail wear. The results indicate that the poor lubrication condition at the rail gauge face is the main cause of severe side wear of the high rail, and the use of wheel profiles with low conicity and the loss of profile curvature of low rail profiles caused by the untimely maintenance also increase the side wear of the high rail. The service life of the high rail could be greatly extended by some countermeasures, such as repairing the worn shape of the low rail and repairing/improving the wayside lubrication system.

Investigation on Tooth Surface Wear of Cycloid Drives Considering Tooth Profile Modifications

Cycloid drives are widely used in various mechanical systems due to their high reduction ratio, compact size, and high efficiency. Tooth surface wear is a major problem that affects the reliability and durability of cycloid drives. However, compared to the research on the wear of involute gears, the prediction of tooth surface wear in cycloid drives is relatively limited and less extensive. To fill this gap, the theoretical wear model of the tooth surface of cycloid-pin gear pairs is proposed based on the Hertz contact theory and Archard’s formula, with consideration of tooth profile modifications. Firstly, the loaded tooth contact analysis model is established to determine the relative sliding velocity and tooth contact stress. Secondly, the calculation steps of single tooth surface wear are presented within one gear mesh cycle. With this, the effects of the tooth profile modifications, the operating conditions such as output torque, input speed, and the assembly eccentricity on the wear depth within one gear mesh cycle are investigated. This study gives a deeper understanding of the tooth surface wear mechanisms of cycloid drives and could be employed to assist gear design and to improve the wear resistance.

FEM wear prediction of ceramic hip replacement bearings under dynamic edge loading conditions

Hard-on-Hard hip implants, specifically ceramic tribo-pair, have produced the highest in-vivo wear resistance, biocompatibility, superior corrosion resistance, and high fracture toughness. However, this ceramic tribo-pair suffers from edge loading, sharply increasing wear and accelerating early implant failures due to micro-separation. Even though in-vitro studies have tested the occurrence of wear due to dynamic edge loading, the Finite Element Method (FEM) gives the advantage of accurately estimating the wear, minimizing the experimental time and cost. A new fundamental FEM model is developed to predict wear for ceramic hip replacement bearings under dynamic edge loading conditions for a fixed separation and fixed inclination angle. The model is directly validated with the existing hip simulator data up to 3 million cycles in terms of wear depth, wear scar and volumetric wear rate. The results from the model show that the accuracy in wear prediction was more than 98% for the wear depth and volumetric wear rate for the dynamic edge loading condition. A stripe wear scar is captured, depicting the edge loading conditions. The developed model from this study can predict wear under pure standard and dynamic edge loading conditions.

A novel deep learning method with partly explainable: Intelligent milling tool wear prediction model based on transformer informed physics

With the trend of lightweight in the field of intelligent electric vehicles and 3C, the demand for high precision machining of aluminum alloy parts is growing. And tool condition monitoring (TCM) is very important for quality control of parts, so intelligent high-accuracy wear prediction of aluminum alloy high precision machining tools has great industrial application value at present and in the future. This paper presents a novel TCM model (Conv-PhyFormer) of Transformer with physics informed. The model has excellent ability to capture short-term and long-term dependencies from nonlinear cutting time series data when there are few training samples. The embedded hard physical constraint and soft physical constraint in the model make the model partially interpretable. Soft physical constraint in the form of one-dimensional causal convolution can help the proposed model better learn the local context. Hard physical constraint in the form of the mathematical equation representing cutting physical knowledge are embedded, thus the model does not need to learn this knowledge from time series data from scratch. A large number of analysis results of aluminum alloy machining experimental data show that the proposed Conv-PhyFormer has significantly superior prediction accuracy and robustness compared with the current three popular deep learning models for TCM. Embedded soft and hard physical constraints can significantly reduce the training epochs of Transformer prediction model.

A method for calculating the real contact area of blade-disk structure combined with truncation theory between rough surfaces

The contact characteristics of the blade-disk structure are difficult to predict due to complex working conditions. This paper presented an improved model to predict its real contact area. Firstly, the deformation state and load of the asperities with different radii on the deterministic rough surface are determined based on improved truncation theory. Subsequently, the real contact area of the blade-disk structure is obtained by combined with the finite element method at a lower computational cost. Finally, the application of the proposed method in the wear prediction of dovetail blade-disk structures is demonstrated, which proves that considering roughness is significant for wear prediction. The proposed method has great potential in predicting the wear characteristics of complex systems under real working conditions.

Dynamics and Wear Prediction of Mechanisms Considering Multiple Clearances and Coatings

The increase in joint clearance and system vibration caused by joint wear are important factors that may lead to the failure of the mechanism and a decrease in the motion accuracy of the actuator. In this paper, to investigate the impact of different numbers of spherical joint clearances on the dynamics performance of the mechanism and explore effective measures to reduce the vibration of the mechanism caused by multiple joint clearances, on the basis of the previous research on the mechanism dynamics when there is clearance in a single spherical joint, an improved contact force model is used to study the effects of coating, no coating, and different clearance quantities on the dynamics of a mechanism containing multiple spherical joint clearances. Moreover, the joint reaction forces and contact-impact forces under different working conditions are calculated. In order to reduce the difficulty of calculating the joint wear, an approximate calculation method for contact area is proposed, and an improved Archard wear model is employed to calculate and analyze the changes in wear amount under the influence of both spherical joint clearance and coating.

Wear Prediction of Curved Switch Rail in High-Speed Turnout and Influence of Wheel and Rail Wear on Vehicle Dynamic Performance

Wear Prediction of Curved Switch Rail in High-Speed Turnout and Influence of Wheel and Rail Wear on Vehicle Dynamic Performance

A hybrid-driven probabilistic state space model for tool wear monitoring

Tool wear monitoring (TWM) is essential to improve product quality and maintain machining safety. Within recent years, numerous techniques have been developed for TWM, including data-driven methods and physics-based models. However, traditional data-driven methods are highly dependent on measured data. The physics-based models are difficult to model complex machining processes. To solve these problems, a hybrid-driven probabilistic state space model is proposed to improve the accuracy and robustness of TWM. In this work, a sensitive feature extraction scheme is first constructed to eliminate interferences and redundancies for improving monitoring efficiency. Subsequently, the Gaussian process is innovatively developed to integrate the data mining and physical model from a probabilistic perspective. On this basis, the particle filter is then utilized to estimate the model parameters and estimate tool wear conditions. Finally, an adaptive wear state recognition method is further established for the predictive maintenance of cutting tools. The practical data obtained from milling experiments are used to validate the effectiveness of the proposed methodology. The results show that the proposed hybrid-driven method effectively improves tool wear prediction accuracy to over 97.7%, while the constructed probabilistic model successfully evaluates the prediction results with over 95% confidence. Therefore, the developed methodology may provide a promising way for tool health management to improve economic efficiency and maintain machining safety.

Establishment of high-temperature fretting wear prediction model for metal rubber with complex helical network structure

In this study, a high-temperature fretting wear prediction model was established for metal rubber, which was then validated through experiments. Initially, a three-dimensional numerical model of metal rubber was reconstructed, and statistical distribution of contact points and types of curved wires inside the virtual component was achieved. Meanwhile, a high-temperature fretting wear evolution model for curved wires was constructed by introducing a temperature function f(T), considering material thermal expansion, softening, and oxidation. Subsequently, the model was incorporated into the macro model, establishing a cross-scale predictive model for wear from micro-coil elements to component wear. Experimental validation demonstrated good prediction performance of the model, with a maximum wear mass prediction error of 14.81% and prediction errors below 10% for 80,000 cycles.

Multi-objective optimization based on machine learning and non-dominated sorting genetic algorithm for surface roughness and tool wear in Ti6Al4V turning

Titanium alloys are notoriously difficult to machine. They are used in the manufacture of various types of lightweight components. It is therefore important to improve their machinability and thus achieve sustainability in machining such alloys, by selecting appropriate influential factors: cutting parameters, tool material, geometric form, coolant types, and hybrid machining methods, to deliver efficient output. Nowadays, meta-heuristic algorithms effectively solve multi-objective optimization in machining problems instead of single-objective one. Along with that, the mathematical predictive models used for single-objective optimization are gradually being replaced by machine learning algorithms, which are highly robust and efficient in terms of prediction performance. Therefore, this work addresses the prediction and optimization of average surface roughness (Ra) and tool wear (VB) in Ti6Al4V alloy turning, using a WC tool coated by chemical vapor deposition (CVD) and physical vapor deposition (PVD), with dry machining. We apply a two-pronged approach combining machine learning (ML) and Non-Dominated Sorting Genetic Algorithm (NSGA-II), to model and optimize Ra and VB. The four ML models – Linear Regression (LIN), Support Vector Machine Regression (SVR), Extreme Gradient Boosting (XGB), and Artificial Neural Network (ANN) – are used to predict Ra and VB. The input variables of the turning process – feed rate, depth of cut, cutting speed, cutting time, and tool materials – are the major factors affecting surface quality and tool wear. By the error metrics such as root mean squared error (RMSE), mean absolute error (MAE), and coefficient of determination (R2), ANN is found to offer the best predictive performance. An ML and NSGA-II-based approach is then employed for multi-objective optimization of cutting parameters in Ti6Al4V turning. Fifty Pareto solutions are identified in the range of Ra and VB between (1.332 and 1.441 µm) and (0.100 and 0.125 mm), respectively. In this work, the Pareto solutions are selected based on their ranked performances. This aligns with the decision criterion employed to select the most robust cutting parameters. The definitive optimal Ra and VB are obtained by formulating a robust decisive multi-criterion function which integrates performance, preferred decision criterion, and trustworthiness. Finally, this produces the optimal solution for Ra and VB − 1.439 µm and 0.100 mm, respectively. Experimental validation confirms that the final optimum solution is within the acceptable range.

Analysis of air foil thrust bearings with annular top foil including wear prediction, Part II: Experimental investigations

Considering air foil thrust bearings, usually a multi-pad design is used. Here, an alternative thrust bearing design with an annular-shaped top foil geometry is considered and investigated experimentally. For the experiments, a dedicated and specially designed high-precision test rig is used. The newly designed test rig allows high-speed measurements of different interesting bearing parameters for thrust loads up to 100N and rotor speeds up to 120krpm. In this work, a special focus concerns the wear occurring in the coating layer of the top foil. It will be shown that wear is significantly influencing the performance of this bearing type. Also, a comparison between the experimentally determined load capacity, power loss, and disk temperature with simulated data generated with a multi-physical TEHD model is given.

Prediction of Tool Wear Rate and Tool Wear during Dry Orthogonal Cutting of Inconel 718

A new prediction method was proposed based on the positive feedback relationship between tool geometry and tool wear rate. Dry orthogonal cutting of Inconel 718 was used as a case study. Firstly, tool wear rate models and a tool wear prediction flowchart were proposed. Secondly, the evolution of the tool geometry during tool wear was analyzed considering the combined effect of tool crater wear and tool flank wear. Thirdly, the evolution of the cutting temperature, normal stress and tool–chip relative sliding velocity on the tool wear surface was studied, the evolution of the tool wear rate during tool wear was revealed. Finally, the evolution of the tool geometry and tool wear rate during tool wear were applied to the tool wear prediction method to accurately predict the tool wear. The prediction error of KT is less than 15% in comparison with the experimental results. The tool wear prediction method in this paper is helpful to improve the prediction accuracy of tool crater wear.

Dynamic computational wear model of PEEK-on-XLPE bearing couple in total hip replacements.

Understanding wear mechanisms is a key factor to prevent primary failures causing revision surgery in total hip replacement (THR) applications. This study introduces a wear prediction model of (Polyetheretherketone) PEEK-on-XLPE (cross-linked polyethylene) bearing couple utilized to investigate the wear mechanism under 3D-gait cycle loading over 5 million cycles (Mc). A 32-mm PEEK femoral head and 4-mm thick XLPE bearing liner with a 3-mm PEEK shell are modeled in a 3D explicit finite element modeling (FEM) program. The volumetric and linear wear rates of XLPE liner per every million cycles were predicted as 1.965 mm3/Mc, and 0.0032 mm/Mc respectively. These results are consistent with the literature. PEEK-on-XLPE bearing couple exhibits a promising wear performance used in THR application. The wear pattern evolution of the model is similar to that of conventional polyethylene liners. Therefore, PEEK could be proposed as an alternative material to the CoCr head, especially used in XLPE-bearing couples. The wear prediction model could be utilized to improve the design parameters with the aim of prolonging the life span of hip implants.

Development of research and prediction of wear of plasma coatings from nickel-clad graphite powder

Development of research and prediction of wear of plasma coatings from nickel-clad graphite powder

The multi-sensor-based measurement of machining signals and data fusion to develop predictive tool wear models for TiAlN-PVD coated carbide inserts during end milling of Inconel 617

Machining signals from the Force dynamometer, Acoustic Emission (AE), and Accelerometer are acquired and fused to develop Machine Learning (ML) models for tool wear monitoring of TiAlN-PVD coated carbide inserts. Milling experiments were performed on Inconel 617 with varied process parameter combinations until the tool flank wear met the failure criterion. Support Vector Regression, Random Forest Regression, and Long Short-Term Memory models are developed and compared based on a combination of sensor data fusion for tool wear predictions. It is observed that the Random Forest Regression approach can predict the tool wear with 94% accuracy compared to Support Vector Regression (85%) and Long Short-Term Memory (84%) models while using three-sensor data fusion. Further, the two-sensor data combination was used to test the relative efficacy of all the three developed machine learning tool wear models and found that the force dynamometer and the AE sensor fared better for Random Forest Regression and Long Short-Term Memory models in comparison to Support Vector Regression. For Support Vector Regression-based tool wear predictive models, force dynamometer and accelerometer data fusion performed better.

A steps-ahead tool wear prediction method based on support vector regression and particle filtering

This paper develops a steps-ahead tool wear prediction method based on particle filtering and support vector regression. A degradation phase classification method is presented based on clustering algorithm and support vector machine. The support vector regression models are established to achieve the mapping between degradation features and the flank tool wear values. In online prediction, the measured signals are input into the SVM model to judge the current phase of the tool, and the particle filtering algorithm is used for online steps-ahead prediction of the features. The prediction of tool wear can be obtained by inputting the feature prediction results into the SVR models. The experimental tool wear dataset is introduced as an application example to test the proposed method. The results demonstrate the effectiveness of the proposed method, and the comparison with other tool wear prediction shows the advantage of the proposed method in prediction accuracy.

Tool wear prediction in milling CFRP with different fiber orientations based on multi-channel 1DCNN-LSTM

Tool wear prediction in milling CFRP with different fiber orientations based on multi-channel 1DCNN-LSTM