A New Approach to Spatial Tool Wear Analysis and Monitoring

Čerče, Luka; Pušavec, Franci; Kopač, Janez

doi:10.5545/sv-jme.2015.2512

Cited by 26 publications

(13 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Normally, tool wear can cause major failure of the cutting tool, which may result in wastage of a product, slow operation, and a lack of productivity. Identifying the wear status of cutting tool is especially important in evaluating product accuracy [4] and [5].…”

Section: Introductionmentioning

confidence: 99%

Deep Stacked Auto-Encoder Network Based Tool Wear Monitoring in the Face Milling Process

Nguyen

Pham

2020

SV-JME

View full text Add to dashboard Cite

Tool wear identification plays an important role in improving product quality and productivity in the manufacturing industry. The actual tool wear status with input cutting parameters may cause different levels of spindle vibration during the machining process. This research proposes an architecture comprising a deep learning network (DLN) to identify the actual wear state of machining tool. Firstly, data on spindle vibration signals are obtained from an acceleration sensor. The data are then pre-processed using the fast Fourier transform (FFT) method to reveal the relevant outstanding features in the frequency domain. Finally, the DLN is constructed based on stacked auto-encoders (SAE) and softmax, which is trained with the input data on the vibration features of the respective tool wear state. This DLN architecture is then used to identify the actual wear statuses of machining tool. The experimental results from the collected data show that the proposed DLN architecture is capable of identifying actual tool wear with high accuracy.Highlights • An expert technique for tool wear monitoring based on an experimental dataset is explored. • The feature values with respect to the wear status of cutting tools are extracted and analyzed. • The effects of a proposed deep learning network architecture for identifying the different tool wear statuses are considered. • A patterns prediction method is compared and developed.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Stacked Auto-Encoder Network Based Tool Wear Monitoring in the Face Milling Process

Nguyen

Pham

2020

SV-JME

View full text Add to dashboard Cite

show abstract

“…A great temperature difference can result in thermal stress on the sample's outer surface [25]. The mean crack length value varied from 42.2 to 111.2 µm in length depending on the cycle number.…”

Section: Metallographic Studymentioning

confidence: 99%

Thermally-Induced Crack Evaluation in H13 Tool Steel

Abdulhadi

Ahmad

Ismail

et al. 2017

Metals

View full text Add to dashboard Cite

This study reported the effect of thermal wear on cylindrical tool steel (AISI H13) under aluminum die-casting conditions. The AISIH13 steels were immersed in the molten aluminum alloy at 700 • C before water-quenching at room temperature. The process involved an alternating heating and cooling of each sample for a period of 24 s. The design of the immersion test apparatus stylistically simulated aluminum alloy dies casting conditions. The testing phase was performed at 1850, 3000, and 5000 cycles. The samples were subjected to visual inspection after each phase of testing, before being examined for metallographic studies, surface crack measurement, and hardness characteristics. Furthermore, the samples were segmented and examined under optical and Scanning Electron Microscopy (SEM). The areas around the crack zones were additionally examined under Energy Dispersive X-ray Spectroscopy (EDXS). The crack's maximum length and Vickers hardness profiles were obtained; and from the metallographic study, an increase in the number of cycles during the testing phase resulted in an increase in the surface crack formation; suggesting an increase in the thermal stress at higher cycle numbers. The crack length of Region I (spherically shaped) was about 47 to 127 µm, with a high oxygen content that was analyzed within 140 µm from the surface of the sample. At 700 • C, there is a formation of aluminum oxides, which was in contact with the surface of the H13 sample. These stresses propagate the thermal wear crack length into the tool material of spherically shaped Region I and cylindrically shape Region II, while hardness parameters presented a different observation. The crack length of Region I was about 32% higher than the crack length of Region II.

show abstract

“…A direct measurement of tool wear on machine tools using laser displacement sensor is presented [3]. A frequent approach used in laboratories is to attach vibration, acoustic emission and dynamometer sensors to the machine and then monitor the signals obtained.…”

Section: Introductionmentioning

confidence: 99%

Integration of Pre-Simulation and Sensorless Monitoring for Smart Mould Machining

Kim¹

2016

Int. j. simul. model.

View full text Add to dashboard Cite

A smart mould machining system, which detects machining status and controls the CNC through presimulation and real-time data, was developed. Pre-simulation predicts cutting forces, and inserts the best feed rate and virtual load on each line of the NC data. The active feed rate reduced machining by up to 36 %, without increasing the maximum cutting forces. The actual cutting load was computed from spindle load data and a friction load compensation algorithm. Collision and tool wear were detected by comparing the actual and virtual load, with the time synchronised by using tool position data. The system machined an automotive grill mould cavity for 34 hours without the supervision of a worker, because pre-simulation had stabilized the milling process and monitoring would have stopped the process if the actual load was different to the virtual load. Pre-simulation has been verified by thousands of mould makers and integrated with sensorless monitoring in an open CNC.

show abstract

A New Approach to Spatial Tool Wear Analysis and Monitoring

Cited by 26 publications

References 17 publications

Deep Stacked Auto-Encoder Network Based Tool Wear Monitoring in the Face Milling Process

Deep Stacked Auto-Encoder Network Based Tool Wear Monitoring in the Face Milling Process

Thermally-Induced Crack Evaluation in H13 Tool Steel

Integration of Pre-Simulation and Sensorless Monitoring for Smart Mould Machining

Contact Info

Product

Resources

About