Transient Detection Data Acquisition for Industrial Assets

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional methods for collecting and processing industrial asset data are inadequate for monitoring fast-developing transients due to low acquisition frequency, leading to excessive network traffic, computational resource issues, and potential loss of transient data, especially when dealing with a large number of assets.

Innovation Solution

The system dynamically adjusts data collection by determining the minimum number of sensor tags needed to detect valid operating state transitions, reading full or minimal subsets of sensor values based on transient periods, and implementing a catch-up mechanism to ensure data availability, thereby reducing network traffic and improving server performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is collected at low acquisition frequency from snapshot buffers, then network traffic is reduced and storage requirements are lowered, but transient signals cannot be detected reliably

Engineering Contradiction:
Improvetransient detection reliabilityVSAvoiddata volume
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

The system dynamically adjusts data collection frequency based on detected transient conditions. During normal operation, data is collected at low frequency to reduce network traffic and storage requirements. When a transient condition is detected, the system automatically increases acquisition frequency to capture the transient signal reliably, then returns to low-frequency collection after the transient ends.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes the acquisition frequency parameter from a fixed low value to a variable value that adapts based on transient detection needs. This allows the system to maintain high reliability for transient detection only when necessary, rather than continuously operating at high frequency which would generate excessive data volume.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If data is collected at high acquisition frequency for all sensors, then transient signals are captured reliably, but network traffic increases substantially and server resources are overwhelmed

Engineering Contradiction:
Improvetransient detection reliabilityVSAvoidsystem efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system extracts and processes only the minimal subset of sensor data necessary for transient detection. Rather than collecting high-frequency data from all sensors continuously, the system identifies and collects data only from sensors relevant to transient conditions, removing unnecessary data collection overhead and improving system efficiency.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system applies partial action by collecting high-frequency data only when and where necessary for transient detection, rather than applying full data collection across all sensors at all times. This selective approach maintains transient detection reliability while avoiding the inefficiency of universal high-frequency collection.

Inventive Principle:
Principle #16Partial or excessive action

3Loss of information

If full sensor data sets are collected continuously, then complete information is available for analysis, but storage requirements and computational resources increase significantly

Engineering Contradiction:
Improveinformation completenessVSAvoidstorage requirements
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The system applies different data collection strategies to different sensors based on their relevance to transient detection. Critical sensors maintain high-frequency collection to preserve information completeness, while non-critical sensors operate at lower frequencies or are sampled selectively, reducing overall storage requirements while maintaining necessary information quality.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system segments the sensor data collection process into critical and non-critical components. By dividing the data collection strategy, the system preserves complete information for transient-relevant sensors while reducing data volume from other sensors, thereby balancing information completeness with storage efficiency.

Inventive Principle:
Principle #1Segmentation

4Speed

If data collection frequency is increased to capture fast-developing transients, then transient detection capability improves, but network traffic and computational time increase

Engineering Contradiction:
Improvetransient response speedVSAvoidnetwork and computational resources
Core Design Contradiction:
SpeedVSLoss of energy

Solution Approach 1:

The system uses periodic low-frequency data collection during normal operation and switches to high-frequency periodic collection only when transient conditions are detected. This periodic adjustment of acquisition frequency maintains fast transient response capability when needed while minimizing network traffic and computational resource consumption during stable operation.

Inventive Principle:
Principle #19Periodic action

Data Source

PatentEP3164819B1Acquisition of high frequency data in transient detection
Publication Date: 2022.10.26 GENERAL ELECTRIC CO
  • EP3164819B1 patent drawingFigure 1
  • EP3164819B1 patent drawingFigure 2

AI summary

Approaches are provided where a group of historical data relating to the operation of an industrial asset are collected on a time interval. It is determined whether the collected group of historical data corresponds to a transitory period of the industrial asset. When the collected group of historical data corresponds to the transitory period a full set of sensor values is read and made available to an analytics engine. When the collected group of historical data does not correspond to the transitory period, a minimal subset of sensor values necessary to detect a start or an end of the transitory period is read and stored.