A Mechanical Fault Diagnosis and Condition Monitoring Method Based on Optimized Fault Characteristic Spectrum
A technology of fault characteristic frequency and fault characteristics, applied in the field of mechanical equipment health monitoring and intelligent operation and maintenance, can solve the problems of lack of intelligent optimization mechanism, weak generalization ability, poor generalization ability, etc., to achieve automatic determination and clarity , Good technical effect, accurate online status monitoring effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0042] In this embodiment, the implementation steps of the envelope demodulation technology are mainly described, specifically as follows:
[0043] The collected original discrete vibration signal or the vibration signal x∈R after signal processing and denoising N×1 (N is the number of sampling points of the signal, and the symbol in bold indicates that it is a vector), and the envelope signal E=|x[n]+jH(x[n])| is obtained by demodulating the Hilbert transform H( ), for the packet Get the square envelope signal SE=E2 after each element of the envelope signal is squared, then apply the fast Fourier transform fft ( ) to the square envelope signal SE processing to obtain the square envelope spectrum SES=fft (SE) (without pair The spectrum obtained by applying the fast Fourier transform directly to the square of the envelope signal is called the envelope spectrum). For the fault signal, the fault type can be judged by identifying the fault characteristic frequency on the square e...
Embodiment 2
[0045] In this embodiment, the implementation steps of a method for optimizing the fault characteristic spectrum based on linear optimal separation are mainly described, as follows:
[0046] The basic idea of this method is to construct a hyperplane in a high-dimensional space to achieve the optimal linear separation of the normalized square envelope spectra of healthy and fault states; The normal vector of the plane) is interpreted as an optimized square envelope spectrum, which is further used for mechanical condition monitoring and fault diagnosis.
[0047] step 1:
[0048] Signal preprocessing, the original vibration signal or the vibration signal x after signal processing and denoising can obtain the square envelope spectrum signal SES (or envelope spectrum signal, and the envelope spectrum signal can also achieve follow-up technology through envelope demodulation technology. Take the square envelope spectrum signal as an example), and then normalize the square envelop...
Embodiment 3
[0081] In this embodiment, the implementation steps of an online mechanical state monitoring and fault diagnosis method based on 3-dimensional optimized fault characteristic spectrum are mainly described. Real-time vibration signals of operating equipment for status monitoring and fault diagnosis. The detailed technical solution roadmap of this method is as follows: figure 1 As shown, the specific steps are as follows:
[0082] step 1:
[0083] Generating Normalized Squared Envelope Spectrum NSES in Healthy State H . When the mechanical equipment is working in a healthy state, a section of vibration signal is collected to generate a normalized square envelope spectrum in a healthy state;
[0084] Step 2:
[0085] Generate the normalized squared envelope spectrum NSES under the initial "abnormal state" F . When starting to monitor the health status of the equipment, the vibration signal collected initially is considered to be in an "abnormal state", and the corresponding...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More