Search results

Conventional wear models cannot satisfy the requirements of electrical contact wear simulation. Therefore, this study aims to establish a novel wear simulation model that considered the influence of thermal-stress-wear interaction to achieve high accuracy under various current conditions, especially high current.

The proposed electrical contact wear model was established by combining oxidation theory and the modified Archard wear model. The wear subroutine was written in FORTRAN, and adaptive mesh technology was used to update the wear depth. The simulation results were compared with the experimental results and the typically used stress-wear model. The temperature of the contact surface, distribution of the wear depth and evolution of the wear rate were analyzed.

With the increase in the current flow, the linear relationship between the wear depth and time changed to the parabola. Electrical contact wear occurred in two stages, namely, acceleration and stability stages. In the acceleration stage, the wear rate increased continuously because of the influence of material hardness reduction and oxidation loss.

In previous wear simulation models, the influence of multiple physical fields in friction and wear has been typically ignored. In this study, the oxidation loss during electrical contact wear was considered, and the thermo-stress-wear complete coupling method was used to analyze the wear process.

The purpose of this paper is building the regression model related to tool wear, and the regression model is used to identify the state of tool wear.

In this paper, genetic programming (GP), which is originally used to solve the symbolic regression problem, is used to build the regression model related to tool wear with the strong regression ability. GP is improved in genetic operation and weighted matrix. The performance of GP is verified in the tool vibration, force and acoustic emission data provided by 2010 prognostics health management.

In result, the regression model discovered by GP can identify the state of tool wear. Compared to other regression algorithms, e.g. support vector regression and polynomial regression, the identification of GP is more precise.

The regression models built in this paper can only make an assessment of the current wear state with current signals of tool. It cannot predict or estimate the tool wear after the current state. In addition, the generalization of model has some limitations. The performance of models is just proved in the signals from the same type of tools and under the same work condition, and different tools and different work conditions may have influences on the performance of models.

In this study, the discovered regression model can identify the state of tool wear precisely, and the identification performances of model applied in other tools are also excellent. It can provide a significant information about the health of tool, so the tools can be replaced or repaired in time, and the loss caused by tool damage can be avoided.

The increasing demands of high-speed railway transportation aggravate the wheel and rail surface wear. It is of great significance to repair the worn wheel timely by predicting the wheel and rail surface wear, which will improve both the service life of the wheel and rail and the safe operation of the train. The purpose of this study is to propose a new prediction method of wheel tread wear, which can provide some reference for selecting proper re-profiling period of wheel.

The standard and worn wheel profiles were first matched with the standard 60N rail profile, and then the wheel/rail finite element models (FEMs) were established for elastic-plastic contact calculation. A calculation method of the friction work was proposed based on contact analysis. Afterwards, a simplified method for calculating wheel tread wear was presented and the wear with different running mileages was predicted.

The wheel tread wear increased the relative displacement and friction of contact spots. There was obvious fluctuation in the wheel tread friction work curve of the worn model. The wear patterns predicted in the present study were in accordance with the actual situation, especially in the worn model.

In summary, the simplified method based on FEM presented in this paper could effectively calculate wheel tread wear and predict the wear patterns. It would provide valuable clews for the wheel repair work.

The purpose of this paper is to adapt teachable machine as a web-based tool for recognition of wear pattern and type of wear by training a convolutional neural network (CNN) model. This helps to monitor the health of the lubricated system as a part of condition monitoring.

Ferrography technique is used for analysis of wear particles. It helps monitor the condition of lubricated mechanical system. In present paper, CNN model is developed for identifying the type of wear particles coming out of Gearbox system using teachable machine.

From the experimentation, it has been observed that the wear severity index has been increased due to increase in wear particle concentration. CNN model has achieved an accuracy of 95.4% to recognize five categories of wear particles.

Teachable machine is generally used for the prediction of images, gestures and sound features. An attempt is made to apply this model for micro and nano wear particles to classify them based on their characteristics.

Excessive tool wear is responsible for damage or breakage of the tool, workpiece, or machining center. Thus, it is crucial to examine tool conditions during the machining process to improve its useful functional life and the surface quality of the final product. AI-based tool wear prediction techniques have proven to be effective in estimating the Remaining Useful Life (RUL) of the cutting tool. However, the model prediction needs improvement in terms of accuracy.

This paper represents a methodology of fusing a feature selection technique along with state-of-the-art deep learning models. The authors have used NASA milling data sets along with vibration signals for tool wear prediction and performance analysis in 15 different fault scenarios. Multiple steps are used for the feature selection and ranking. Different Long Short-Term Memory (LSTM) approaches are used to improve the overall prediction accuracy of the model for tool wear prediction. LSTM models' performance is evaluated using R-square, Mean Absolute Error (MAE), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) parameters.

The R-square accuracy of the hybrid model is consistently high and has low MAE, MAPE and RMSE values. The average R-square score values for LSTM, Bidirection, Encoder–Decoder and Hybrid LSTM are 80.43, 84.74, 94.20 and 97.85%, respectively, and corresponding average MAPE values are 23.46, 22.200, 9.5739 and 6.2124%. The hybrid model shows high accuracy as compared to the remaining LSTM models.

The low variance, Spearman Correlation Coefficient and Random Forest Regression methods are used to select the most significant feature vectors for training the miscellaneous LSTM model versions and highlight the best approach. The selected features pass to different LSTM models like Bidirectional, Encoder–Decoder and Hybrid LSTM for tool wear prediction. The Hybrid LSTM approach shows a significant improvement in tool wear prediction.

Remanufacturing of worn blades with various defects normally requires processes such as scanning, regenerating a geometrical reference model, additive manufacturing (AM) through laser cladding, adaptive machining and polishing and quality inspection. Unlike the manufacturing process of a new part, the most difficult problem for remanufacturing such a complex surface part is that the reference model adaptive to the worn part is no longer available or useful. The worn parts may suffer from geometrical deformation, distortion and other defects because of the effects of harsh operating conditions, thereby making their original computer aided design (CAD) models inadequate for the repair process. This paper aims to regenerate the geometric models for the worn parts, which is a key issue for implementing AM to build up the parts and adaptive machining to reform the parts. Unlike straight blades with similar cross sections, the tip geometry of the worn tip of a twist blade needs to be regenerated by a different method.

This paper proposes a surface extension algorithm for the reconstruction of a twist blade tip through the extremum parameterization of a B-spline basis function. Based on the cross sections of the scanned worn blade model, the given control points and knot vectors are firstly reconstructed into a B-spline curve D. After the extremum of each control point is calculated by extremum parameterization of a B-spline basis function, the unknown control points are calculated by substituting the extremum into the curve D. Once all control points are determined, the B-spline surface of the worn blade tip can be regenerated. Finally, the extension algorithm is implemented and validated with several examples.

The proposed algorithm was implemented and verified through the exampled blades. Through the extension algorithm, the tip geometry of the worn tip of a twist blade can be regenerated. This method solved a key problem for the repair of a twist blade tip. It provides an appropriate reference model for repairing worn blade tips through AM to build up the blade tip and adaptive machining/polishing processes to reform the blade geometry.

The extension errors for different repair models are compared and analyzed. The authors found that there are several factors affecting the accuracy of the regenerated model. When the cross-section interval and the extension length are set properly, the restoration accuracy for the blade tip can be improved, which is acceptable for the repairing.

The lack of a reference geometric model for worn blades is a significant problem when implementing blade repair through AM and adaptive machining processes. Because the geometric reference model is unavailable for the repair process, reconstruction of the geometry of a worn blade tip is the first crucial step. The authors proposed a surface extension algorithm for the reconstruction of a twist blade tip. Through the implementation of the proposed algorithm, the blade tip model can be regenerated.

Remanufacturing of worn blades with various defects is highly demeaned for the aerospace enterprises considering sustainable development. Unlike straight blades, repair of twist blades encountered a very difficult problem because the geometric reference model is unavailable for the repair processes. This paper proposed a different method to generate the reference model for the repair of a twist blade tip. With this model, repair of twist blades can be implemented through AM to build up the blade tip and adaptive machining to subtract the extra material.

The authors proposed a surface extension algorithm to reconstruct the geometric model for repair of twist blades.

To avoid braking accidents caused by excessive wear of brake pad, this study aims to achieve online prediction of brake pad wear life (BPWL).

A simulated braking test bench for automobile disc brake was used. The correlation and mechanism between the three braking condition parameters of initial braking speed, braking pressure and initial braking temperature and the tribological performance were analyzed. The different artificial neural network (ANN) models of wear loss were discussed. Genetic algorithm was used to optimize the ANN model. The structure scheme of the online prediction system of BPWL was discussed and completed.

The results showed that the braking conditions were positively correlated with the wear loss, but negatively correlated with the friction coefficient. The prediction accuracy of back propagation (BP) ANN model was higher. The model was optimized by genetic algorithm, and the average deviation of prediction results was 4.67%. By constructing the online monitoring system of automobile braking conditions, the online prediction of BPWL based on the ANN model could be realized.

The research results not only have important theoretical significance for the study of BPWL but also have practical value for guiding the maintenance and replacement of automobile brake pads and avoiding the occurrence of braking accidents.

Quantitative fretting wear prediction is of practical significance for industrial components. This study aims to establish a fretting wear model considering the formation of tribolayers and provide better fretting wear prediction.

Based on the characteristics for the formation of tribolayers, the ratio of fretting amplitude to nominal contact area length in the fretting direction is used to characterize their formation and contribution to the wear volume. The wear volume is then associated with the product of the friction energy and the ratio of fretting amplitude to nominal contact area length.

Better prediction in the wear volume can be achieved with the proposed fretting wear model by taking the formation of tribolayers into consideration.

The contribution of the formation of tribolayers to the wear volume is considered in the model and better prediction can be achieved.

The peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-01-2023-0004/

The purpose of this paper is to identify smaller wear particles and improve the calculation speed, identify more abrasive particles and promote industrial applications.

This paper studies a new intelligent recognition method for equipment wear debris based on the YOLO V5S model released in June 2020. Nearly 800 ferrography pictures, 23 types of wear debris, about 5,000 wear debris were used to train and test the model. The new lightweight approach of wear debris recognition can be implemented in rapidly and automatically and also provide for the recognition of wear debris in the field of online wear monitoring.

An intelligent recognition method of wear debris in ferrography image based on the YOLO V5S model was designed. After the training, the GIoU values of the model converged steadily at about 0.02. The overall precision rate and recall rate reached 0.4 and 0.5, respectively. The overall MAP value of each type of wear debris was 40.5, which was close to the official recognition level of YOLO V5S in the MS COCO competition. The practicality of the model was approved. The intelligent recognition method of wear debris based on the YOLO V5S model can effectively reduce the sensitivity of wear debris size. It also has a good recognition effect on wear debris in different sizes and different scales. Compared with YOLOV. YOLOV, Mask R-CNN and other algorithms%2C, the intelligent recognition method based on the YOLO V5S model, have shown their own advantages in terms of the recognition effect of wear debris%2C the operation speed and the size of weight files. It also provides a new function for implementing accurate recognition of wear debris images collected by online and independent ferrography analysis devices.

To the best of the authors’ knowledge, the intelligent identification of wear debris based on the YOLO V5S network is proposed for the first time, and a large number of wear debris images are verified and applied.

This study aims to investigate the effect of time-varying passenger flow on the wheel wear of metro vehicles to provide a more accurate model for predicting wheel wear and a new idea for reducing wheel wear.

Sectional passage flow data were collected from an operational metro line. A wheel wear simulation based on time-varying passenger flow was performed via the SIMPACK software to obtain the worn wheel profile and wear distribution. The simulation involves the following models: vehicle system dynamics model, wheel-track rolling contact model, wheel wear model and variable load application model. Later, the simulation results were compared with those obtained under the traditional constant load condition and the measured wear data.

For different distances traveled by the metro vehicle, the simulated wheel profile and wear distribution under the variable load remained closer to the measurements than those obtained under the constant load. As the distance traveled increased, the depth and position of maximum wear and wear growth rate under the variable load tended to approach the corresponding measured values. In contrast, the simulation results under the constant load differed greatly from the measured values. This suggests that the model accuracy under the variable load was significantly improved and the simulation results can offer a more accurate basis for wear prediction.

These results will help to predict wheel wear more accurately and provide a new idea for simulating wheel wear of metro vehicles. At the same time, measures for reducing wheel wear were discussed from the perspective of passenger flow changes.

Existing research on the wheel wear of metro vehicles is mainly based on the constant load condition, which is quite different from the variable load condition where the passenger flow in real vehicles varies over time. A method of simulating wheel wear based on time-varying load is proposed in this paper. The proposed method shows a great improvement in simulation accuracy compared to traditional methods and can provide a more accurate basis for wear prediction and wheel repair.