Spectral Peak Compression for Machine Health Data Transmission
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high bandwidth requirements for transmitting machine spectral data over networks, particularly in manufacturing facilities with numerous wireless vibration monitoring devices, lead to network congestion and slow data transfer speeds, especially when large amounts of data need to be transmitted simultaneously for analysis.
Innovation Solution
A data compression process that identifies and stores only the N number of highest-amplitude spectral peaks and calculates corresponding RMS values, along with optional spectral side values, to reduce the data transmitted while maintaining analysis details, thereby reducing network traffic and storage needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complete spectral data is transmitted over the network, then analysis accuracy is maintained, but network bandwidth consumption increases
Solution Approach 1:
The patent extracts only the most significant spectral peaks (top N peaks by amplitude) from the complete spectral data for transmission. This selective extraction maintains the essential diagnostic information needed for machine health analysis while eliminating redundant data points, thereby reducing network bandwidth consumption without significantly compromising analysis accuracy.
Solution Approach 2:
The patent applies different quality levels to different parts of the spectral data. High-amplitude peaks that contain critical diagnostic information are preserved in full detail, while low-amplitude regions are represented by RMS values or omitted entirely. This local differentiation optimizes the balance between data fidelity and transmission efficiency.
2Loss of information
If all spectral data points are transmitted, then complete information is available for analysis, but transmission time increases
Solution Approach 1:
The patent extracts only the essential spectral features (N highest amplitude peaks and their surrounding RMS values) for transmission. This extraction maintains the critical information needed for diagnostic analysis while dramatically reducing the total data volume, thereby shortening transmission time without significant loss of diagnostic capability.
Solution Approach 2:
The patent transmits a partial representation of the spectral data that includes all diagnostically relevant information (peaks and RMS values) while omitting redundant low-amplitude regions. This partial transmission approach achieves sufficient information completeness for analysis while minimizing transmission time.
3Measurement precision
If high-resolution spectral data is transmitted, then diagnostic detail is preserved, but storage requirements increase
Solution Approach 1:
The patent extracts and stores only the essential diagnostic elements (N spectral peaks with their amplitude and frequency values, plus RMS values for surrounding regions) rather than storing complete high-resolution spectral data. This extraction preserves the diagnostic detail needed for analysis while significantly reducing storage requirements.
Solution Approach 2:
The patent applies different storage quality levels to different spectral regions. High-amplitude peak regions are stored with full resolution to preserve diagnostic detail, while low-amplitude regions are represented by aggregated RMS values or omitted, thereby optimizing the balance between diagnostic detail and storage efficiency.
4Productivity
If multiple devices transmit simultaneous spectral data, then comprehensive monitoring is achieved, but network congestion increases
Solution Approach 1:
The patent enables each monitoring device to extract and transmit only its essential spectral features (N peaks and RMS values) rather than complete spectral data. This extraction approach allows multiple devices to simultaneously transmit compressed data, achieving comprehensive facility-wide monitoring while keeping individual and aggregate network traffic volumes manageable.
Data Source
AI summary
A data compression process reduces the amount of machine spectral data transmitted over a network while maintaining the details of spectral peaks used for machine health analysis. The data compression process also provides for the calculation of various types of spectral parameters, such as spectral band parameters, with negligible loss of accuracy.