Sparse Time Synchronous Averaging for Rotating Machine Fault Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing time synchronous averaging method for rotating machine fault diagnosis is ineffective against rotating frequency fluctuations and requires long signal lengths, limiting its ability to extract fault feature components efficiently.
Innovation Solution
A fault detection method based on sparse time synchronous averaging that constructs a sparse weighting matrix using rotating frequency information, performs sparse time synchronous averaging, and calculates the STSA_CI index from the sparse frequency spectrum and reconstructed time signal to diagnose faults in rotating machines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional time synchronous averaging method is used, then periodic components can be extracted and signal-to-noise ratio can be improved, but the method has poor performance against rotating frequency fluctuations and requires long signal lengths
Solution Approach 1:
The patent applies parameter changes by transforming the traditional time synchronous averaging method into a sparse time synchronous averaging method. This involves changing the mathematical parameters of the averaging process, using sparse representation theory to reformulate the signal processing approach. The transformation allows the method to achieve better fault feature extraction with shorter signal lengths by optimizing the mathematical parameters of the averaging operation.
Solution Approach 2:
The patent implements dynamics by making the averaging window adaptive rather than fixed. The sparse time synchronous averaging method dynamically adjusts to rotating frequency fluctuations by using sparse representation to identify and track the actual fault-related frequencies in real-time. This dynamic adaptation allows the method to maintain high extraction accuracy even when operating conditions change, eliminating the need for long signal lengths to accommodate frequency variations.
2Productivity
If traditional time synchronous averaging method is used, then calculation speed is fast, but the method can only extract same frequency components and has poor performance against frequency fluctuations
Solution Approach 1:
The patent applies universality by designing a sparse time synchronous averaging method that can extract multiple different frequency components simultaneously. The sparse representation framework allows the method to identify and extract various fault-related frequencies (such as gear mesh frequencies, bearing characteristic frequencies, and their harmonics) in a single processing operation. This multi-functional capability enables the method to adapt to different fault types and frequency conditions while maintaining fast calculation speed.
Solution Approach 2:
The patent uses parameter changes to enhance the extraction capability for different frequency components. By reformulating the averaging process using sparse representation theory, the method changes the mathematical parameters to allow selective extraction of multiple frequency components. The sparse optimization process automatically identifies which frequency components are present in the signal, enabling versatile fault feature extraction without sacrificing computational efficiency.
Data Source
AI summary
A fault detection method for a rotating machine based on sparse time synchronous averaging is disclosed, and the method includes: collecting a vibration signal and a rotating frequency or rotating frequency pulse signal of the rotating machine and performing analog-to-digital conversion to obtain the vibration signal and rotating speed information by a sensor; according to the type and number of detection components in the rotating machine, constructing a component-aware comb vector g based on the vibration signal and rotating speed information, wherein the type includes the gear, rotor and bearing; constructing a quasi-time synchronous average vector w based on the component-aware comb vector g; constructing a sparse time synchronous averaging model F by using the quasi-time synchronous average vector w; solving the sparse time synchronous averaging model F with an optimization solution algorithm to obtain a sparse spectrum and a reconstruction time signal.