Prediction of surface roughness in end face milling based on Gaussian process regression and cause analysis considering tool vibration

Zhang, Guojun; Jian, Li; Cao, Yuan; Huang, Yu; Shao, Xinyu; Li, Mingzhen

doi:10.1007/s00170-014-6232-6

Cited by 39 publications

(12 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Additionally, Khasawneh [23] had combined supervised machine learning with topological data analysis to obtain a descriptor of the process which can detect chatter in turning. And Zhang et al [24] used Gaussian process regression (GPR) for modelling and predicting surface roughness in end face milling with accuracy of 84%. Furthermore, Aguiar et al [25] have developed a neural network using a multisensor method to predict the final roughness on the grinded workpiece with 70% success rate.…”

Section: Parameters Selection Based On Aimentioning

confidence: 99%

Roundness prediction in centreless grinding using physics-enhanced machine learning techniques

Safarzadeh

Leonesio

Bianchi

et al. 2020

Int J Adv Manuf Technol

View full text Add to dashboard Cite

This work proposes a model for suggesting optimal process configuration in plunge centreless grinding operations. Seven different approaches were implemented and compared: first principles model, neural network model with one hidden layer, support vector regression model with polynomial kernel function, Gaussian process regression model and hybrid versions of those three models. The first approach is based on an enhancement of the well-known numerical process simulation of geometrical instability. The model takes into account raw workpiece profile and possible wheel-workpiece loss of contact, which introduces an inherent limitation on the resulting profile waviness. Physical models, because of epistemic errors due to neglected or oversimplified functional relationships, can be too approximated for being considered in industrial applications. Moreover, in deterministic models, uncertainties affecting the various parameters are not explicitly considered. Complexity in centreless grinding models arises from phenomena like contact length dependency on local compliance, contact force and grinding wheel roughness, unpredicted material properties of the grinding wheel and workpiece, precision of the manual setup done by the operator, wheel wear and nature of wheel wear. In order to improve the overall model prediction accuracy and allow automated continuous learning, several machine learning techniques have been investigated: a Bayesian regularized neural network, an SVR model and a GPR model. To exploit the a priori knowledge embedded in physical models, hybrid models are proposed, where neural network, SVR and GPR models are fed by the nominal process parameters enriched with the roundness predicted by the first principle model. Those hybrid models result in an improved prediction capability.

show abstract

Section: Parameters Selection Based On Aimentioning

confidence: 99%

Roundness prediction in centreless grinding using physics-enhanced machine learning techniques

Safarzadeh

Leonesio

Bianchi

et al. 2020

Int J Adv Manuf Technol

View full text Add to dashboard Cite

show abstract

“…It was proved that, for a multivariable machining problem, gray relationalbased orthogonal array Taguchi method is best suited. The cutting parameters and the vibration signature are mapped using Gaussian process regression (GPR) for estimation of the surface texture tolerance by (Zhang et al 2014). Face milling operation is performed on C45E4 steel.…”

Section: Literature Reviewmentioning

confidence: 99%

Development of surface texture evaluation system for highly sparse data-driven machining domains

Raju

Ramesh

Varma

2020

International Journal of Computer Integrated Manufacturing

View full text Add to dashboard Cite

Dimensional, geometrical and surface texture tolerances are significant issues to be addressed by manufacturing industries. Dimensional and geometrical tolerance estimation systems are used by several machining industries, whereas surface texture tolerance estimation systems are rare. In general-purpose machines, several machining operations are performed and huge machining data are required for the development of surface texture tolerance estimation model. The necessity for sparse data modeling is the need of the hour and such modeling techniques reduce costly trial and error approaches. Shoulder milling operations are performed on mild steel workpieces. Experimentation is performed at diverse cutting conditions, and surface texture tolerances of the machined components are measured. Big data computation is being accomplished by contemporary tools compared to sparse data evaluation. The novelty of this work is to develop a system capable of surface texture tolerance evaluation from highly sparse learning data, as big data generation involves cost and time for experimentation. Flower Pollination Algorithm (FPA) models are developed for the highly sparse data for surface roughness estimation. Two different FPA techniques, namely, Maximum (Max) and Average (Avg) are used, and a comparison between the methods mentioned above is made. An operator-friendly adaptive performance enhancement system is developed to evaluate the corresponding surface texture tolerance for the given operating parameters.

show abstract

“…The influences of three levels of spindle-attributed forced vibrations along with axial cutting depth and feed rate were assessed in terms of dimensional accuracy, surface roughness, and tool wear under constant conditions of cutting speed and radial cutting depth. Zhang et al 21 proposed a model for predicting surface roughness during the end face milling process based on Gaussian process regression and cause analysis and considering tool vibration. Dong et al 22 established a generalized dynamic model for spindle vibration to study its distinctive effects on surface topography in different ultra-precision machining processes.…”

Section: Introductionmentioning

confidence: 99%

Modelling and simulation of surface topography machined by peripheral milling considering tool radial runout and axial drift

Chen

Wang

2019

Proceedings of the Institution of Mechanical Engineers, Part B:

View full text Add to dashboard Cite

Based on the Z-map model of a workpiece and the dynamic cutting forces model of peripheral milling in which the regenerative effect of tool radial runout and axial drift are considered, a model for the prediction of surface topography in peripheral milling operations is presented. According to the stability lobe diagram obtained by the zero-order analytical method, the relationship between spindle speed and surface topography, the tool radial runout, and the axial drift following the chatter are studied. The results show that a stable cutting status but a poor surface finish is obtained at the spindle speeds at which the dominant frequency of the milling system is integral multiples of the selected machining frequency, and a stable cutting status with a good surface finish can be obtained near and on the left side of the resonant spindle speeds determined by the predicted stability lobe diagram. The motion equations of any tooth end mill for peripheral milling are established, and these equations are based on the transformation matrix and the vector operation principle of motion-homogeneous coordinates. In addition, the simulation algorithm and the system of surface topography generated in peripheral milling are given based on the Z-map model. Cutting tests are carried out, and good agreement between the measured surface topographies and the topographies predicted by the model in this study is found in terms of their shape, magnitude, feed mark, profile height of cross-section, and surface roughness. The simulation results show that the milling surface roughness increases with the increase in feed per tooth, which further shows that this simulation system has high credibility. Thus, the simulation and experimental results can provide some practical instructions for the actual peripheral milling in determining the optimal machining conditions.

show abstract

Prediction of surface roughness in end face milling based on Gaussian process regression and cause analysis considering tool vibration

Cited by 39 publications

References 34 publications

Roundness prediction in centreless grinding using physics-enhanced machine learning techniques

Roundness prediction in centreless grinding using physics-enhanced machine learning techniques

Development of surface texture evaluation system for highly sparse data-driven machining domains

Modelling and simulation of surface topography machined by peripheral milling considering tool radial runout and axial drift

Contact Info

Product

Resources

About