Vibration Spectrum Windowing for Small-Peak Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing vibration spectra in machine condition monitoring are inefficient, requiring analysts to repeatedly replot data and utilize multiple windows to compare small amplitude peaks and different domains, leading to resource and time inefficiencies.
Innovation Solution
A processor-based system that performs a Fast Fourier Transform on vibration data, generates spectral plots with modifiable windows, allowing users to adjust the window position and scale, and switch between variables like displacement, velocity, or acceleration, using a user interface for interactive analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the analyst amplifies a portion of the vibration spectrum with small peaks, then the small peaks become distinguishable, but the analyst cannot compare the portion to the rest of the spectrum because other peaks are out of range
Solution Approach 1:
The vibration spectrum is divided into a main spectral plot showing the full frequency range and a modifiable window showing an enlarged portion of interest. This segmentation allows simultaneous viewing of both the detailed small peaks in the window and the context of the entire spectrum in the main plot, resolving the contradiction between detecting small peaks and maintaining comparison capability.
Solution Approach 2:
The modifiable window is nested within the spectral plot, creating a hierarchical display structure. The window can be positioned at different locations and scaled independently while remaining part of the overall spectral visualization. This nested structure enables the analyst to examine enlarged portions of the spectrum without losing the ability to compare with the complete spectral context.
2Measurement precision
If the analyst changes the plotted variable to better analyze a particular peak, then the peak characteristics become clearer, but other regions of the plot become indecipherable upon conversion
Solution Approach 1:
Different variables can be plotted in different regions - the main spectral plot displays one variable (e.g., acceleration) while the modifiable window displays another variable (e.g., velocity or displacement). This local differentiation allows optimal visualization of specific peak characteristics in the window without compromising the interpretability of other regions, as each region maintains its own variable context.
Solution Approach 2:
The system adds a dimensional aspect by allowing variable transformation within the modifiable window while maintaining the original variable display in the main plot. This creates an additional analytical dimension where the same frequency data can be viewed in different physical domains (displacement, velocity, acceleration) without losing the original context.
3Measurement precision
If the analyst repeatedly replots the data to compare portions with different properties, then accurate analysis is achieved, but computing resources and analyst time are inefficiently utilized
Solution Approach 1:
The system performs preliminary actions by pre-computing and storing the vibration spectrum data in multiple domains (displacement, velocity, acceleration) and preparing modifiable windows that can be instantly configured. When the analyst needs to compare different portions or variables, the data is already prepared and can be displayed immediately by adjusting window parameters rather than requiring repeated data processing and reploting, significantly improving efficiency while maintaining analytical accuracy.
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
Enables efficient comparative analysis of vibration spectra by allowing dynamic adjustment of plots, reducing the need for repeated data replotting and improving resource utilization, facilitating more effective machine condition monitoring.
Implementation Method 1
perform a Fast Fourier Transform on the machine vibration data to generate a vibration spectrum that defines an amplitude of a first variable as a function of frequency
Data Source
AI summary
While monitoring the condition of a machine, vibration data is often collected for analysis by an experienced analyst. Systems and methods for analyzing vibration spectra associated with machine condition monitoring are disclosed herein. A system may be configured to collect vibration data from one or more vibration sensors, generate a vibration spectrum of the vibration data, and generate a spectral plot of the vibration spectrum. The system may receive a selection of a region of the spectral plot and generate a modifiable window of the vibration spectrum that is embedded within the spectral plot. The system may display a set of graphing tools along with the modifiable window that enable a user to make modifications to the window. The system may detect the modifications and update the modifiable window accordingly.