In-situ Wear Monitoring: An Experimental Investigation of Acoustic Emission During Thread Forming

Wirtz, Sebastian Felix; Demmerling, Anna Lena; Söffker, Dirk

doi:10.12783/shm2017/13987

Cited by 2 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nukman, 2012;Kosaraju, Anne, & Popuri, 2013;Srinivasan, Bhinge, & Dornfeld, 2016). Meanwhile, AE signals can also be applied for distinction of MWF during threading (Wirtz, Demmering, & Söffker, 2017) besides the well-known tapping torque test according to ASTM D5619 and advanced approaches (Demmerling & Söffker, 2020). Kmeans clustering was applied to classify the AE energy in different frequency bands which was a feature to distinguish different MWF qualities (Wirtz et al 2017) least one of the layers.…”

Section: Introductionmentioning

confidence: 99%

“…Meanwhile, AE signals can also be applied for distinction of MWF during threading (Wirtz, Demmering, & Söffker, 2017) besides the well-known tapping torque test according to ASTM D5619 and advanced approaches (Demmerling & Söffker, 2020). Kmeans clustering was applied to classify the AE energy in different frequency bands which was a feature to distinguish different MWF qualities (Wirtz et al 2017) least one of the layers. Compare with other neural networks, the innovation of convolutional neural network is based on the ability to automatically learn a large number of filters in parallel to the trained model under the constraints of specific predictive modeling problem, such as image classification (Brownlee, 2018.).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Metalworking Fluid Classification Based on Acoustic Emission Signals and Convolutional Neural Network

Wei

Demmerling²,

Söffker³

2021

PHME_CONF

View full text Add to dashboard Cite

Acoustic emission (AE) which describes the transient stresswaves generated by the rapid release of energy from solidsources has been widely used in nondestructive testing(NDT) of materials and structures especially in healthmonitoring. As a class of deep neural networks,convolutional neural network (CNN) has applications inmany fields. Several investigations have been conducted onthe application of CNN in feature learning and faultdiagnosis and prognosis. Metalworking fluids (MWF) play asignificant role in manufacturing processes. By reducingfriction between tool and workpiece, the heat generation inmetalworking process is affected. Thread forming is atransformative manufacturing process for generating threadsin ductile materials. As the thread geometry is manufacturedby cold forming of the material, lubricating properties of theMWF strongly effect tool wear and workpiece quality. Upto now, there are only a few papers on MWF classificationusing the process variables like torque or released AE. Inthis contribution, a novel approach combining AE signalsand CNN is raised for MWF classification. A tribometer isused to carry out thread forming trials under well-controlledexperimental conditions. AE measurements are conducted incontext of thread forming. The AE signals are divided intosuitable samples and CNN is applied as classifier. Theresults of MWF classification show that the new approachcould distinguish different types of MWF.

show abstract