Hello, I am trying to write a program in Matlab to estimate the remaining useful life of a bearing without data of a previous failure. Do you think this is possible? I was thinking to use the amplitude of the ball pass frequency as an indicator but I am not sure how to do so.
Does someone know a paper (or even a Matlab program) which deals with this issue?
I'm not sure this can even be done computationally. The various frequencies associated with rolling element bearings (such as ball pass frequency) are normally only elevated when a defect develops in the ball or race. Also, rolling element bearings are susceptible to vibration in the unloaded state. There have been many examples of damage occurring during transport. This effect is often referred to as false brinelling.
Have a look at https://mae.ufl.edu/nkim/Papers/paper103.pdf
https://www.hindawi.com/journals/sv/2018/4068431/ and Article Remaining Useful Life Prediction of Rolling Bearings Using a...
Muhammad Ali Thank you for recommending the paper. I am trying to understand equation 1 of your first paper. However, I have some difficulties to understand the idea behind the bins.
I have ten measurements of the acceleration signal with a length of 5040 samples. The acceleration signal is almost a sin wave. If I want to put the acceleration data into the bins, the bin would be evenly distributed. Could someone tell me the mistake I am doing or explain the idea behind the bins in other words? I would really appreciate that!
- Badges
- Science topic