In-process Tool Wear Prediction System Based on Machine Learning Techniques and Force Analysis

Gouarir, Amine; Martínez-Arellano, Giovanna; Terrazas, Germán; Benardos, Panorios; Ratchev, Svetan

doi:10.1016/j.procir.2018.08.253

Cited by 148 publications

(51 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As was stated in Section 1, the most adequate method is using analysis of the forces. In the literature, one can find multiple examples which present the correlation between the tool wear and the cutting forces [32][33][34][35][36]. For that reason, the load distribution for a spherical bowl punch was analyzed and the obtained model is presented in Fig.…”

Section: Influence Of the Geometrical Features Of The Punch And Die Omentioning

confidence: 99%

Determination of the effective geometrical features of the piercing punch for polymer composite belts

Wojtkowiak

Talaśka

2019

Int J Adv Manuf Technol

View full text Add to dashboard Cite

Punch optimization is important in improving the technological process of belt perforation. Research about this process can rarely be found in the scientific literature. The objective of this paper is to determine the effective geometry of the piercing punches with a spherical bowl used for polymer composite belts. In order to fulfill this goal, the complex research of the influence of the geometrical features of the piercing punch on the perforation force, the hole quality, and the tool life has been conducted. For derivation of the proposed model, the authors have used the combination of analytical and FEM analyses, along with the experimental tests validation. Based on the results, there are clear correlations between the depth of the spherical bowl punch and the perforation force or the hole diameter deviation. By analyzing the ratio of the perforation and transverse force, we are able to evaluate how the tool life will change for different sets of geometrical parameters. All these correlations have been used to derive the indicators of the perforation force, tool life, and hole quality. Based on those, the optimization function has been specified. Finding the minimum value of this function makes it possible to define the optimal tool geometry. The design of effective tools for belt perforation with diameters D = 5, 6, 8, and 10 mm has been presented, but the obtained characteristic makes it possible to find an optimal solution for any diameter which lays in the range from 3 to 10 mm. Since the model has been built based on the indicators, it is easy to adjust it to match any specific requirements of the belt manufacturers, which makes it useful for tool designers.

show abstract

Section: Influence Of the Geometrical Features Of The Punch And Die Omentioning

confidence: 99%

Determination of the effective geometrical features of the piercing punch for polymer composite belts

Wojtkowiak

Talaśka

2019

Int J Adv Manuf Technol

View full text Add to dashboard Cite

show abstract

“…As described before, the first one reads sequences of triplets, which are passed on to the Classifier module. Subsequences of these, i.e., arrays of consecutive triplets, are randomly sampled without repetition and passed, in turns, to the GASF component [39], which is in charge of producing polar coordinate images that feed into the Convolutional Neural Network (CNN) Model component implemented in Tensorflow. Then, the CNN Model analyses these images in order to report a class that reflects the current stage of the tool wear.…”

Section: Computing Architecturementioning

confidence: 99%

“…For instance, for the Break-in class, it is required to sample from only one layer, whereas for the Failure class, it is required to sample from three different layers. Thus, after systematically sampling arrays of triplets from the training set, these are used as input to the GASF component [39], which performs two main tasks. First, it generates an image of 256 × 256 pixels for each of the force components of the sequence; second, it compiles and creates a three-channel object with the created images.…”

Section: Model Trainingmentioning

confidence: 99%

Online Tool Wear Classification during Dry Machining Using Real Time Cutting Force Measurements and a CNN Approach

Terrazas

Martínez-Arellano

Benardos

et al. 2018

JMMP

View full text Add to dashboard Cite

The new generation of ICT solutions applied to the monitoring, adaptation, simulation and optimisation of factories are key enabling technologies for a new level of manufacturing capability and adaptability in the context of Industry 4.0. Given the advances in sensor technologies, factories, as well as machine tools can now be sensorised, and the vast amount of data generated can be exploited by intelligent information processing techniques such as machine learning. This paper presents an online tool wear classification system built in terms of a monitoring infrastructure, dedicated to perform dry milling on steel while capturing force signals, and a computing architecture, assembled for the assessment of the flank wear based on deep learning. In particular, this approach demonstrates that a big data analytics method for classification applied to large volumes of continuously-acquired force signals generated at high speed during milling responds sufficiently well when used as an indicator of the different stages of tool wear. This research presents the design, development and deployment of the system components and an overall evaluation that involves machining experiments, data collection, training and validation, which, as a whole, has shown an accuracy of 78%.

show abstract

“…More recently, machine learning techniques were also presented to determine the tool wear evolution. For example, training artificial neural networks and variants in order to determine the tool wear from tool micrography [16,17], or to predict the tool wear from condition monitoring variables such as cutting forces [18]. F. W. Taylor presented a relationship between the lifetime of a cutting tool (T) and the cutting speed (v c ) [19], which has been extensively discussed and used in numerous studies [6,7,11,20,21].…”

Section: Introductionmentioning

confidence: 99%

An Analytic Approach to the Cox Proportional Hazards Model for Estimating the Lifespan of Cutting Tools

Equeter¹,

Ducobu²,

Rivière-Lorphèvre³

et al. 2020

JMMP

View full text Add to dashboard Cite

The machining industry raises an ever-growing concern for the significant cost of cutting tools in the production process of mechanical parts, with a focus on the replacement policy of these inserts. While an early maintenance induces lower tool return on investment, scraps and inherent costs stem from late replacement. The framework of this paper is the attempt to predict the tool inserts Mean Up Time, based solely on the value of a cutting parameter (the cutting speed in this particular turning application). More specifically, the use of the Cox Proportional Hazards (PH) Model for this prediction is demonstrated. The main contribution of this paper is the analytic approach that was conducted about the relevance on data transformation prior to using the Cox PH Model. It is shown that the logarithm of the cutting speed is analytically much more relevant in the prediction of the Mean Up Time through the Cox PH model than the raw cutting speed value. The paper also covers a numerical validation designed to show and discuss the benefits of this data transformation and the overall interest of the Cox PH model for the lifetime prognosis. This methodology, however, necessitates the knowledge of an analytical law linking the covariate to the Mean Up Time. It also shows how the necessary data for the numerical experiment was obtained through a gamma process simulating the degradation of cutting inserts. The results of this paper are expected to help manufacturers in the assessment of tool lifespan.

show abstract

In-process Tool Wear Prediction System Based on Machine Learning Techniques and Force Analysis

Cited by 148 publications

References 14 publications

Determination of the effective geometrical features of the piercing punch for polymer composite belts

Determination of the effective geometrical features of the piercing punch for polymer composite belts

Online Tool Wear Classification during Dry Machining Using Real Time Cutting Force Measurements and a CNN Approach

An Analytic Approach to the Cox Proportional Hazards Model for Estimating the Lifespan of Cutting Tools

Contact Info

Product

Resources

About