A Feature Extraction Method of Blade Unbalance Fault Electrical Signal Based on Hilbert Transform
A balance fault and feature extraction technology, applied in static/dynamic balance test, machine/structural component test, instrument, etc., can solve the problem that electrical signals are difficult to accurately extract blade unbalance fault features, etc., to achieve algorithm or decision-making Easy application and high stability effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
specific Embodiment approach 1
[0022] Specific implementation mode one: combine figure 1 Description of this embodiment, a method for extracting electrical signal features of blade unbalance fault based on Hilbert transform, the specific process is as follows:
[0023] Step 1: Hilbert transform the input electrical signal to obtain the instantaneous frequency signal f corresponding to the electrical signal e (t) and the corresponding instantaneous frequency f of bearing rotation under ideal conditions r (t);
[0024] Step 2: Calculate the time series S corresponding to the peak value of the electrical signal peak (k);
[0025] Step 3: Calculate the mean frequency f corresponding to each electrical signal cycle e,mean (i) and the mean frequency f corresponding to each bearing rotation cycle r,mean (j);
[0026] Step 4: For the f obtained in Step 3 e,mean (i) and f r,mean (j) Perform cubic spline interpolation to obtain an interpolated signal with the same sampling frequency as the original signal a...
specific Embodiment approach 2
[0045] Specific embodiment 2: This embodiment is a further description of specific embodiment 1. In step 1, Hilbert transform is performed on the input monitoring electrical signal, and the instantaneous frequency signal f corresponding to the electrical signal is obtained. e (t) and the corresponding instantaneous frequency f of bearing rotation under ideal conditions r The process of (t) is:
[0046] Set the original input electrical signal as x(t), time t=1,2,...,N,
[0047] Step a: Perform discrete convolution on the original electrical signal x(t) to obtain its Hilbert transform y(t), as shown in the following formula:
[0048]
[0049] Step b: Calculate the envelope amplitude a(t) of the analytical signal c(t)+jy(t) of the original electrical signal, as shown in the following formula:
[0050]
[0051] Step c: Calculate the phase angle θ(t) of the analytical signal c(t)+jy(t) and the instantaneous frequency f corresponding to the electrical signale (t), as shown ...
specific Embodiment approach 3
[0055] Specific embodiment three: This embodiment is a further description of specific embodiment one. In step two, the time series S corresponding to the peak value of the original electrical signal is calculated. peak The process of (k) is:
[0056] Step a: search for all peak points in the original electrical signal;
[0057] Step b: Calculate the time point information corresponding to all the peak points in step a, and obtain the time series S peak (k), where k=1,2,...,K, k represents the kth peak point, and K represents the total number of peak points contained in the original electrical signal.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


