Variable-Resolution Circular Buffers for Machine Data Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine health monitoring systems face limitations in compressing discrete parameter data points, such as temperature or pressure, and fail to provide information on predictability and variability, leading to inefficient storage and communication of machine operational data.
Innovation Solution
A data acquisition system with a primary circular buffer and variable resolution circular buffers that compress parameter data points, storing calculated compressed data points along with variance and mean square error, allowing for efficient storage and communication of machine operational data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data compression is applied to reduce memory requirements and bandwidth, then storage capacity and communication efficiency are improved, but information on predictability and variability of parameter data points is lost
Solution Approach 1:
The patent segments the data storage system into multiple circular buffers with different compression levels. The primary circular buffer stores uncompressed parameter data points, while secondary circular buffers store compressed data points. This segmentation allows the system to preserve both detailed information and compressed representations, preventing information loss while achieving memory reduction.
Solution Approach 2:
The patent changes the compression parameter by calculating compressed data points using variance and mean square error metrics. By monitoring these parameters, the system can determine when to compress data and how much compression to apply, thereby preserving predictability and variability information while reducing storage requirements.
2Quantity of substance
If data compression is applied to reduce bandwidth for data transmission, then communication efficiency is improved, but diagnostic capability is reduced
Solution Approach 1:
The patent segments transmitted data into multiple resolution levels. The system can transmit highly compressed data points with variance and mean square error metrics for routine monitoring, while retaining the capability to transmit uncompressed detailed data when enhanced diagnostic capability is required, thus balancing bandwidth efficiency with diagnostic reliability.
Solution Approach 2:
The patent implements feedback mechanisms where compressed data points include variance and mean square error metrics. These feedback parameters allow the receiving system to assess data quality and reliability, enabling informed decisions about whether additional diagnostic data transmission is necessary.
3Measurement precision
If high-resolution data is stored to maintain diagnostic accuracy, then measurement precision is improved, but memory requirements increase
Solution Approach 1:
The patent divides the storage system into primary and secondary circular buffers. The primary buffer maintains high-resolution uncompressed data for accurate diagnostics, while secondary buffers store compressed data for long-term storage. This segmentation allows the system to maintain diagnostic accuracy for recent data while reducing overall memory requirements through compression of historical data.
Solution Approach 2:
The patent implements dynamic data management where the system can adjust between storing compressed and uncompressed data based on diagnostic requirements. When enhanced diagnostics are needed, the system can access uncompressed data from the primary buffer; otherwise, it uses compressed data from secondary buffers, optimizing the balance between precision and memory usage.
Data Source
AI summary
A machine data acquisition system is disclosed. The data acquisition system may include a sensor disposed on a machine and configured to produce a signal indicative of an operational parameter of the machine. The system may also include a data processor and controller configured to convert the signal into a plurality of parameter data points. The system may further include a primary data circular buffer disposed on the machine and configured to store the plurality of parameter data points. Additionally, a variable resolution circular buffer may be disposed on the machine and configured to store a compressed parameter data point calculated from the plurality of parameter data points.