Machining process monitoring with vibration signal based manifold learning

2020 ◽  
Vol 148 (4) ◽  
pp. 2766-2766
Author(s):  
Jing Wang ◽  
Mingxin Hui ◽  
Bin Liu ◽  
Xun Wang ◽  
Xiaobin Cheng ◽  
...  
Keyword(s):  
Process Monitoring ◽  
Manifold Learning ◽  
Vibration Signal ◽  
Machining Process

Study on Improving Accuracy in Robotic Milling of Aluminum Alloy

2018 ◽  
Author(s):  
Shaochun Sui ◽  
Kai Guo ◽  
Jie Sun ◽  
Yiran Zang
Keyword(s):  
Aluminum Alloy ◽  
Industrial Robot ◽  
Vibration Signal ◽  
Industrial Robots ◽  
Machining Process ◽  
Cnc Machine ◽  
Machined Surface ◽  
Improving Accuracy ◽  
High Level ◽  
Monolithic Components

Nowadays, the application of using industrial robots in manufacture is a diminutive due to its own low rigidity and low stiffness. This leads to high level of vibrations that limits the quality and the precision of the workpiece. So they are usually used for welding, grinding and paint shop. However, the potential of industrial robot applications in machining has be realized. The volume of monolithic components is large and there are many issues in machining process such as geometric tolerance and quality of machined surface. In such cases the traditional CNC machine is replaced by industrial robots, which will reduce the production cost, reduce labor and increase the efficiency. In this paper, the milling experiment of 7050-T7451 aeronautical aluminum alloy was carried out by using industrial robot KR210 R2700. In addition, the experiment was employed to study the influence of milling speed, feed-rate, cutting depth and cutting width on vibrations, surface roughness was also measured to evaluate the machining quality. Besides, the axis of angle was changed which led to the different industrial robot’s postures. The vibration signal of different postures was acquired, which was used to analysis the optimal workspace of industrial robot. The best process parameters were obtained, which will play a guiding significance on the actual production.

Method for Evaluating Machining Toolholder-Spindle Unbalance Due to Chips at the Toolholder Interface

2010 ◽  
Author(s):  
John Agapiou
Keyword(s):  
Real Time ◽  
Process Monitoring ◽  
Diagnostic System ◽  
Machining Process ◽  
Monitoring Method ◽  
Spindle System ◽  
On Line

Machining process monitoring method is developed for detecting and diagnosis of the presence of chips at the toolholder-spindle interface. Although toolholders can be simply balanced before they are placed in the spindle, there can be some balancing problems remaining when one or more loose machining chips are attached at the toolholder-spindle interface(s) during a tool change. A method is developed by considering the natural and geometric unbalances of the toolholder-spindle system combined with an analysis of the toolholder tilt due to the presence of loose machining chips around the spindle. The method can be integrated on-line as a real-time expert diagnostic system for toolholder tilt due to the presence of loose machining chips at the spindle nose. The expert diagnostic system makes intelligent decisions on toolholder unbalance and concerns with chips at the interface that result in unwanted tilting and vibrations. The tool unbalance algorithm was able to monitor the toolholder tilting according to the results of this study.

scholarly journals Planetary Gearbox Fault Diagnosis Using Envelope Manifold Demodulation

2016 ◽  
Vol 2016 ◽  
pp. 1-13 ◽  
Author(s):  
Weigang Wen ◽  
Robert X. Gao ◽  
Weidong Cheng
Keyword(s):  
Fault Diagnosis ◽  
Frequency Band ◽  
Manifold Learning ◽  
Planetary Gear ◽  
Vibration Signal ◽  
Planetary Gearbox ◽  
Characteristic Extraction ◽  
Gear Fault ◽  
Planet Gear ◽  
Fault Characteristic

The important issue in planetary gear fault diagnosis is to extract the dependable fault characteristics from the noisy vibration signal of planetary gearbox. To address this critical problem, an envelope manifold demodulation method is proposed for planetary gear fault detection in the paper. This method combines complex wavelet, manifold learning, and frequency spectrogram to implement planetary gear fault characteristic extraction. The vibration signal of planetary gear is demodulated by wavelet enveloping. The envelope energy is adopted as an indicator to select meshing frequency band. Manifold learning is utilized to reduce the effect of noise within meshing frequency band. The fault characteristic frequency of the planetary gear is shown by spectrogram. The planetary gearbox model and test rig are established and experiments with planet gear faults are conducted for verification. All results of experiment analysis demonstrate its effectiveness and reliability.

Milling Process Monitoring Based on Vibration Analysis Using Hilbert-Huang Transform

2018 ◽  
Vol 12 (5) ◽  
pp. 688-698 ◽  
Author(s):  
Agus Susanto ◽  
Chia-Hung Liu ◽  
Keiji Yamada ◽  
Yean-Ren Hwang ◽  
Ryutaro Tanaka ◽  
...  
Keyword(s):  
Signal Processing ◽  
Feature Extraction ◽  
Process Monitoring ◽  
Signal Analysis ◽  
Vibration Analysis ◽  
Milling Process ◽  
Machining Process ◽  
Hilbert Huang Transform ◽  
Thin Walled ◽  
Tool Damage

Vibration analysis is one method of machining process monitoring. The vibration obtained in machining is often nonlinear and of a nonstationary nature. Therefore, an appropriate signal analysis is needed for signal processing and feature extraction. In this research, vibrations obtained in the milling of thin-walled workpieces were analyzed using the Hilbert-Huang transform (HHT). The features obtained by the HHT served as machining-state indicators for machining process monitoring. Experimental results showed the effectiveness of the HHT method for detecting chatter and tool damage.

Smart Machining Process Monitoring Enabled by Contextualized Process Profiles for Synchronization

2020 ◽  
Vol 4 (2) ◽  
pp. 20190040
Author(s):  
Zhigang Wang ◽  
Timothy C. Wagner ◽  
Changsheng Guo
Keyword(s):  
Process Monitoring ◽  
Machining Process ◽  
Smart Machining
Sign in / Sign up

Export Citation Format

Share Document