Engine knocking characteristic frequency extraction method
A technology of feature frequency and extraction method, applied in computer parts, special data processing applications, complex mathematical operations, etc., can solve problems such as energy leakage, redundant signals, human factors, etc., to achieve fast processing speed and good denoising effect. Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0048] In the step S3, the Mallat decomposition method is as follows:
[0049] For any function f(t)∈L 2 (R), if let A j (t)∈V j Represents a scale of 2 j The approximation of the function f(t), D j (t) represents the error of approximation, then:
[0050]
[0051]
[0052] From multiresolution analysis, it can be obtained that:
[0053] A j-1 (t)=A j (t)+D j (t) (3)
[0054] Therefore, the relationship between the scale transform coefficients of the function and the wavelet transform coefficients can be written as:
[0055]
[0056] Then the Mallat decomposition algorithm is:
[0057]
[0058]
[0059] Such as figure 2 As shown, the decomposition process of the Mallat decomposition algorithm automatically produces a pyramid-shaped successive decomposition result.
[0060] The body vibration signal data sequence of a two-stroke aviation kerosene engine under a certain working condition is decomposed into signal components in different frequency domain...
Embodiment 2
[0062] In the step S4, the process of obtaining multi-order intrinsic modal components and residual components by using the empirical mode decomposition method for the low-frequency main body information is as follows:
[0063] Find all the maximum points of the low-frequency main signal sequence s(t), and use the cubic spline interpolation function to fit the upper envelope e of the original data + ;Find all the minimum points of the low-frequency main signal sequence s(t), and use the cubic spline interpolation function to fit the lower envelope e of the data - , the mean value of the upper and lower envelopes is denoted as m, and the low-frequency subject information sequence s(t) is subtracted from the average envelope m to obtain a new data sequence h, as shown below:
[0064] m=(e + +e - ) / 2 (7)
[0065] h=s(t)-m (8)
Embodiment 3
[0067] On the time history of h, the difference between the number of zero-crossing points and the number of extreme points is less than or equal to 1, and in the time domain of the research object, the upper and lower points determined by the cubic spline fitting maximum and minimum points The average value of the envelope is 0.
[0068] Obtain the first-order IMF denoted as C 1 , the C 1 Removed from the low-frequency main signal data s(t) to obtain the difference signal r 1 :
[0069] r 1 =s(t)-C 1 (9)
[0070] will r 1 As a new signal, the IMF component screening process is performed again, the next-order IMF component is separated, and a new difference signal is obtained as a new signal, and the above process is repeated continuously, as shown below:
[0071] r n = r n-1 -C n (10)
[0072] Decomposition process such as image 3 As shown, the empirical mode decomposition process is the new data after subtracting the envelope average from the original data. If...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More