Data driven modeling and estimation of accumulated damage in mining vehicles using on-board sensors

Erik Jakobsson, Erik Frisk, Robert Pettersson, and Mattias Krysander
Submission Type: 
Full Paper
AttachmentSizeTimestamp
phmc_17_011.pdf2.83 MBAugust 15, 2017 - 6:44am

NB: Extended abstract & full paper attached as PDF

Accumulated fatigue damage is estimated for a mine truck load frame. On-board sensor data and a linear auto-regressive model is used to calculate stress levels at a number of locations. Lasso regression and coherence is used to identify the most important on-board sensors, such as accelerometers and gyroscopes. A rain flow counting technique is then applied to calculate an estimate of the accumulated damage during different drive cycles of the machine. The accumulated damage based on accelerometer and gyroscope signals is validated against damage accumulation calculations from a measured strain signal. Different load levels of the truck, velocity and road conditions are also taken into consideration.

Publication Year: 
2017
Publication Volume: 
8
Publication Control Number: 
011
Page Count: 
10
Submission Keywords: 
condition monitoring
system identification
damage accumulation
Submission Topic Areas: 
Data-driven methods for fault detection, diagnosis, and prognosis
Industrial applications
Model-based methods for fault detection, diagnostics, and prognosis
Structural health monitoring
Submitted by: 
  
 
 
 

follow us

PHM Society on Facebook Follow PHM Society on Twitter PHM Society on LinkedIn PHM Society RSS News Feed