Advancements in Online Saw Monitoring for Wood Industry 4.0

 
S
a
w
 
s
c
i
e
n
c
e
:
N
e
w
 
r
e
s
e
a
r
c
h
 
o
n
 
o
n
l
i
n
e
 
s
a
w
m
o
n
i
t
o
r
i
n
g
 
Vahid Nasir, Ph.D. Candidate
Department of Wood Science
 
Online monitoring and industry 4.0
 
 
Artificial intelligence
 
Industry 4.0 within the sawmill context
 
Tool health monitoring
Wear detection
Chatter detection
Surface quality assessment
Energy consumption monitoring
 
Literature overview
 
Little research emphasis on wood primary applications
 extreme cutting conditions
 different saw blade characteristics
 
Lack of studies on waviness monitoring
 focus on surface roughness monitoring
 
 Lack of practicing sophisticated signal processing and machine
learning methods
 
Monitoring essentials
 
 
Monitoring
 
8
 
Feature selection
 
Manual
 
Not necessarily optimized
Can be linked to an optimization
model
 
 
Automatic
 
Black-box approach
 
Case study
 
Monitoring the cutting power and waviness in the circular sawing process of
Douglas-fir wood
 
Data acquisition
 
Waviness
 
Impact of feed and rotation speeds
 
Power
 
Waviness
 
1. Manual feature selection
 
Objectives:
Evaluate the performance of
an acoustic emission sensor
in predicting power and
waviness
Identify the optimal sensory
features to maximize the
performance of a neural
network
 
Signal processing and extracted features
Maximum R
2
 of 0.98 and 0.89
for power consumption and
waviness prediction
 
Results
 
2. Automatic feature selection
 
Objectives:
Evaluate the performance of
an accelerometer sensor in
predicting power and
waviness
Quantify the potential of
using an automatic feature
selection process in
combination to a neural
network
Artificial
neural
network
Power and
waviness
monitoring
 
Automatic selection
Vibration signals
Self-organizing maps
Artificial
neural
network
Power and
waviness
monitoring
 
Results
Vibration signals
Self-organizing maps
Maximum R
2
 of 1 can
be achieved for both
power and waviness
 
What is coming?
 
 
Smart manufacturing
Big datasets
Deep
machine
learning
techniques
 
Conclusion
 
Canadian sawmills could benefit from using sensors and employing
sophisticated signal processing and artificial intelligence methods for
accurate process condition and tool health monitoring
 
 
Thank you!
 
 
 
Vahid.nasir@alumni.ubc.ca
Slide Note
Embed
Share

Explore the latest research on online saw monitoring in the context of Industry 4.0, focusing on tools like cloud computing, smart sensors, and artificial intelligence for enhanced monitoring capabilities in the sawmill industry. Discover the importance of tool health monitoring, wear detection, and energy consumption monitoring within the sawmill operations. Gain insights from literature overview discussing the need for more emphasis on wood cutting conditions and advanced signal processing techniques. Dive into case studies on monitoring cutting power and waviness in sawing processes, highlighting the impact of feed and rotation speeds.


Uploaded on Sep 12, 2024 | 0 Views


Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

E N D

Presentation Transcript


  1. Saw science: Saw science: New research on online saw New research on online saw monitoring monitoring Vahid Nasir, Ph.D. Candidate Department of Wood Science

  2. Online monitoring and industry 4.0

  3. Cloud computing Smart sensors Internet of things Industry 4.0 Automation and Robotics Big data analytics

  4. Artificial intelligence

  5. Industry 4.0 within the sawmill context Tool health monitoring Wear detection Chatter detection Surface quality assessment Energy consumption monitoring

  6. Literature overview Little research emphasis on wood primary applications extreme cutting conditions different saw blade characteristics Lack of studies on waviness monitoring focus on surface roughness monitoring Lack of practicing sophisticated signal processing and machine learning methods

  7. Monitoring essentials

  8. Monitoring Sensor selection Signal processing and feature extraction Artificial Intelligence Feature Selection Acoustic Accelerometer Acoustic emission Current Laser displacement sensor 8

  9. Feature selection Manual Automatic Not necessarily optimized Can be linked to an optimization model Black-box approach

  10. Case study Monitoring the cutting power and waviness in the circular sawing process of Douglas-fir wood

  11. Data acquisition Waviness

  12. Impact of feed and rotation speeds Power Waviness

  13. 1. Manual feature selection Objectives: Evaluate the performance of an acoustic emission sensor in predicting power and waviness Identify the optimal sensory features to maximize the performance of a neural network

  14. Signal processing and extracted features

  15. Results Sensory features Maximum R2 of 0.98 and 0.89 for power consumption and waviness prediction Optimization model (PSO) ANN

  16. 2. Automatic feature selection Objectives: Evaluate the performance of an accelerometer sensor in predicting power and waviness Quantify the potential of using an automatic feature selection process in combination to a neural network

  17. Automatic selection Artificial neural network Power and waviness monitoring Self-organizing maps Vibration signals

  18. Results Maximum R2 of 1 can be achieved for both power and waviness Artificial neural network Power and waviness monitoring Self-organizing maps Vibration signals

  19. What is coming?

  20. Smart manufacturing Deep machine learning techniques Big datasets

  21. Conclusion Canadian sawmills could benefit from using sensors and employing sophisticated signal processing and artificial intelligence methods for accurate process condition and tool health monitoring

  22. Thank you! Vahid.nasir@alumni.ubc.ca

Related


More Related Content

giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#