Lubricating Oil Wear Particle Signal Extraction Using Segmentation Entropy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for extracting wear particle feature signals from lubricating oil using inductive particle detection sensors face challenges in accurately identifying and counting ferromagnetic wear particles due to interference from noise, electrical impulses, and harmonic interferences, leading to distortion and deformation of the signals.
Innovation Solution
A method based on segmentation entropy is employed, involving low-pass filtering, harmonic interference suppression, and adaptive thresholding to segment and extract wear particle feature signals, using a sliding window and normalized segmentation entropy to enhance signal retention and noise reduction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If decomposition methods are used to separate target signal and noise into different frequency bands, then noise reduction is achieved, but signal distortion and deformation occur due to wide frequency band coverage
Solution Approach 1:
The patent segments the signal processing task by dividing the raw signal into multiple fixed-length time-domain sequence segments using a sliding window approach. Each segment is processed independently to calculate segmentation entropy, allowing localized noise reduction while preserving the morphological features of wear particle signals across different time intervals
Solution Approach 2:
The patent transforms the signal from time domain to entropy domain by calculating segmentation entropy for each time-domain segment. This parameter transformation enables the system to identify and remove noise segments based on entropy thresholds while preserving valid wear particle signal segments, achieving noise reduction without signal distortion
2Object-affected harmful factors
If adaptive filtering is used to improve signal-to-noise ratio, then noise is reduced, but signal distortion still occurs and residual random noise remains
Solution Approach 1:
The patent replaces traditional adaptive filtering mechanisms with an entropy-based segmentation approach. Instead of using filter coefficients to attenuate noise frequencies, the system calculates segmentation entropy for each time-domain segment and compares it against thresholds to identify and remove noise segments, eliminating the signal distortion problems inherent in filter-based approaches
3Object-affected harmful factors
If frequency band separation is used for signal extraction, then noise cancellation is achieved, but effectiveness is reduced when target signal frequency covers unwanted components
Solution Approach 1:
The patent changes the domain of analysis from frequency domain to entropy domain. By calculating segmentation entropy for each time-domain segment and comparing against dynamically determined thresholds, the system can reliably identify noise segments regardless of frequency overlaps between target signals and interference, significantly improving algorithm robustness
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method effectively preserves the morphology features of wear particle signals while significantly reducing noise, enabling accurate recognition and extraction of wear particle signals, thus facilitating better wear condition monitoring of mechanical equipment.
Implementation Method 1
An inductive particle detection sensor, based on the principle of inductance, has been widely used in the lubricating oil wear particle monitoring of mechanical equipment
Implementation Method 2
a preprocessed signal may be obtained by performing low-pass filtering and harmonic interference suppression on the raw signal to be processed
Data Source
AI summary
A method for extracting a wear particle feature signal based on segmentation entropy is provided, including obtaining a raw signal to be processed by performing real-time data acquisition using a lubricating oil wear particle monitoring system; obtaining a preprocessed signal by performing low-pass filtering and harmonic interference suppression on the raw signal to be processed; dividing the preprocessed signal into a plurality of time domain sequence segments with a sliding window; calculating segmentation entropy corresponding to each time domain sequence segment, normalizing a segmentation entropy set to obtain normalized segmentation entropy; obtaining an adaptive threshold through curve fitting based on empirical cumulative distribution of normalized segmentation entropy, obtaining a plurality of non-zero discrete time domain signal segments by segmenting the preprocessed signal by the adaptive threshold; and obtaining final extraction results of the wear particle feature signal by excluding residual noise interference through target signal feature recognition indices.