Multi-Sensor Correlation for Industrial Abnormal Condition Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial operations face challenges in effectively aggregating and correlating data from diverse sensors to identify abnormal operating conditions, as existing systems often utilize sensor information solely for component functionality, missing opportunities for trend detection and predictive maintenance.
Innovation Solution
A method and system that receive sensor information from multiple sensors, derive correlations between component and additional sensors, create a baseline signature of normal operating conditions, and identify abnormal conditions by comparing additional sensor data to this signature, sending alerts for potential issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor information is used solely for component functionality, then component operation is maintained, but opportunities for trend detection and predictive maintenance are missed
Solution Approach 1:
The patent applies multi-functionality by enabling sensor information to serve dual purposes: its original component functionality and additional trend detection/predictive maintenance functions. The system processes the same sensor data through multiple analytical pathways, allowing a single sensor input to fulfill multiple operational roles without requiring additional sensors.
Solution Approach 2:
The system performs preliminary action by establishing baseline signatures from historical sensor data before abnormal conditions occur. These baselines are pre-computed and stored, enabling rapid comparison and detection of deviations from normal operation, thus preparing the system in advance for predictive maintenance activities.
2Measurement precision
If correlations are derived from multiple sensor types, then abnormal condition detection is enhanced, but system complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the complex multi-sensor analysis into distinct functional modules: baseline signature generation, correlation derivation, and deviation detection. Each module handles a specific aspect of the analysis, processing sensor data in discrete stages rather than attempting simultaneous comprehensive analysis, thus managing complexity while maintaining detection accuracy.
Solution Approach 2:
The system uses baseline signatures as an intermediary between raw sensor data and abnormal condition detection. These baselines serve as a reference medium that simplifies the comparison process, allowing the system to detect deviations without directly complexly analyzing all sensor correlations in real-time.
3Productivity
If sensor data is repurposed for dual purposes, then predictive maintenance is enhanced, but data processing requirements increase
Solution Approach 1:
The system maintains continuity of useful action by continuously updating baseline signatures and correlations using incoming sensor data without interruption. Rather than batch processing or periodic analysis, the system continuously refines its predictive models using the same sensor stream that monitors component functionality, ensuring productive use of data flow without redundant processing cycles.
Data Source
AI summary
A method for correlating data from sensors includes receiving sensor information from a plurality of sensors of an industrial operation. Sensor information from component sensors is used for functionality of a component of the industrial operation and sensor information from additional sensors monitor conditions of a portion of the industrial operation different from the component. The method includes deriving, using the sensor information, correlations between component sensors and additional sensors and deriving a baseline signature from the sensor information and the correlations. The baseline signature encompasses a range of normal operating conditions. The method includes identifying an abnormal operating condition based on a comparison between additional sensor information and the baseline signature. The sensor information is used differently for functionality of the component than for deriving the correlations and baseline signature and identifying the abnormal operating condition. The method includes sending an alert with the abnormal operating condition.