Field Device Reliability Modeling With Threshold-Based Data Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in assessing the reliability of field devices under variable conditions is the difficulty in conducting laboratory tests due to high reliability and environmental dependency, leading to data scarcity and the need for extensive data storage, which is impractical and inaccurate when reduced.
Innovation Solution
A method and system for assessing reliability using dynamic compression of field device data based on environmental and usage variables, ensuring accuracy by comparing compressed and uncompressed values against predefined thresholds, allowing for efficient storage and reanalysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data from many devices is stored over an extended period to address data scarcity, then reliability assessment accuracy is improved, but data storage requirements increase significantly
Solution Approach 1:
The patent extracts and stores only the most relevant features and patterns from the raw field data, rather than storing all raw data points. This selective extraction maintains the essential information needed for reliability assessment while significantly reducing the storage burden of the massive datasets generated by thousands of devices over extended periods.
Solution Approach 2:
Instead of storing all raw data and processing it later, the patent inverts the approach by first processing and compressing the data into essential features during collection, then storing only these processed features. This reversal of the traditional data storage pipeline achieves space efficiency without sacrificing assessment accuracy.
2Volume of stationary object
If data is reduced to make storage practical, then storage requirements are decreased, but reliability assessment accuracy deteriorates
Solution Approach 1:
The patent applies different processing and compression strategies to different portions of the data based on their importance. Critical features that directly impact reliability assessment are preserved with high fidelity, while less critical data is compressed more aggressively, achieving space efficiency without sacrificing the accuracy of the assessment.
Solution Approach 2:
The patent transforms the raw data into different parameter representations that are more compact yet retain the essential information for reliability assessment. By changing the parameter space and using feature extraction techniques, the system maintains assessment accuracy while reducing storage requirements.
3Reliability
If laboratory tests are conducted to assess device reliability, then reliability data can be obtained, but testing becomes difficult due to high device reliability and environmental dependencies
Solution Approach 1:
Instead of conducting complex laboratory tests on physical devices, the patent creates a virtual model of device reliability by collecting and processing data from multiple field-deployed devices. This digital copy or model of reliability behavior provides the necessary reliability data without requiring physically complex and environmentally sensitive laboratory testing.
Data Source
AI summary
A method for assessing reliability of field device data incudes starting a cycle; receiving reliability-related data; calculating and accumulating a first reliability and/or hazard value over a time span without compression according to a model that uses at least one variable representing the reliability-related data; calculating and accumulating a second reliability and/or hazard value with compression according to the model, which uses the at least one variable representing the reliability-related data, wherein the values of the at least one variable are compressed; comparing the first and second reliability and/or hazard values to a related pre-defined threshold corresponding to an accuracy; when exceeding the threshold, stopping the calculation of the first and second values and determining the time span; and storing the compressed values of the at least one variable; and closing the individual cycle.