Method for rolling bearing fault detection based on enhancing statistical asymmetry
AssigneeRubico AB
InventorsPatrik Paajarvi, James P Leblanc
Life of this patent
Estimating the term…
Abstract
A method for digital processing of vibration signals from rolling bearings in rotating machines is presented, allowing subsequent fault detection with high reliability. A linear, adaptive filter is applied to the acquired vibration signal and iteratively tuned to increase the statistical asymmetry of its output. In this process, the filter removes phase- and amplitude distortion from underlying fault impulses. Furthermore, suppression of sinusoidal disturbances is simultaneously achieved with high robustness to measurement noise. The result is a processed signal from which rolling bearing defects are more easily detected.
Claims5
A method of indicating a fault in a rolling bearing in a rotating machine by improving the quality of a digital vibration signal from the rotating machine through suppression of disturbances and noise, and removal of transfer function distortion, the method comprising the steps of:
- providing a computer platform;
- processing said digital vibration signal using said computer platform implementing a linear, digital, adaptive filter, whose output signal is modelled as a stationary stochastic process, with each sample, a random variable X having a probability density function fx(x) and mean value px;
- adjusting the parameters of said linear, digital, adaptive filter in a control algorithm resulting in an increase of the magnitude of ϕ(X), with ϕ(X) defined as
- ϕ
-
- (
- X
- )
-
- =
- Δ
-
- ∫
- -
- ∞
- ∞
-
- g
-
- (
- x
- )
-
- f
- X
-
- (
- x
- -
- μ
- X
- )
-
-
- dx
- h
-
- (
- X
- )
- ϕ being an indicator of the statistical asymmetry of X;
- g(x) being an arbitrary non-linear and odd function of x and h(X) chosen, dependent on g(x), so that ϕ(αX)=ϕ(X) for any α>0;
- and
- examining a frequency spectrum of the output signal of said linear, digital, adaptive filter;
- comparing values of the frequency spectrum at frequency points corresponding to characteristic frequencies of said rolling bearing;
- and
- wherein a fault in said rolling bearing in said rotating machine is indicated if said values in said frequency spectrum exceed a set value.
Dependencies are inferred from the claim text.