Abnormality Factor Identification Using Operating State Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for isolating abnormality factors in industrial machines face challenges due to the difficulty in selectively responding to specific abnormalities, often requiring multiple sensors and preliminary knowledge that may not be readily available, especially in environments with varying installation conditions and infrequent fault occurrences.
Innovation Solution
An abnormality factor identification apparatus that combines information about machine operating states and sensor signals to calculate operational abnormality levels, generate historical data, and analyze time-series profiles to isolate abnormality factors, allowing for accurate identification with a minimal number of sensors and without relying on preliminary knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sensors are installed to selectively respond to specific abnormalities, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the abnormality detection task by dividing the analysis into multiple operating states (e.g., idle, running, stopping). For each state, separate abnormality level calculations are performed, allowing precise detection without requiring multiple physical sensors. This virtual segmentation replaces physical sensor multiplication with computational segmentation.
Solution Approach 2:
A single sensor is made multi-functional by analyzing its output across different operating states. The same sensor signal serves multiple detection purposes by calculating abnormality levels specific to each operating state, thereby replacing multiple specialized sensors with one universal sensor that performs multiple detection functions through computational differentiation.
2Productivity
If preliminary knowledge about sensor observation values is used to isolate abnormality factors, then productivity is improved, but adaptability deteriorates
Solution Approach 1:
The system collects sensor observation values across multiple operating states and uses this feedback to dynamically calculate abnormality levels. Rather than relying on pre-stored knowledge, the system adapts to each specific operating condition by computing abnormality levels based on actual observed data, enabling both rapid response and environmental adaptability through real-time feedback processing.
Solution Approach 2:
The abnormality level calculation is made dynamic by adjusting the analysis based on the current operating state. The system transitions between different calculation modes depending on whether the machine is idle, running, or stopping, allowing the detection methodology to adapt dynamically to changing conditions rather than relying on static preliminary knowledge.
3Measurement precision
If sensors are installed near target sites to sense abnormalities, then measurement precision is improved, but ease of operation deteriorates
Solution Approach 1:
The patent extracts the abnormality detection function from the physical sensor installation and relocates it to the computational analysis stage. By calculating abnormality levels from sensor data across different operating states rather than installing specialized sensors at each target site, the system removes the complexity of precise sensor placement while maintaining detection precision through information extraction and computational analysis.
Data Source
AI summary
An abnormality factor identification apparatus includes a sensor signal obtaining unit that obtains sensor signals associated with the physical state of a machine, an operating state determination unit that determines operating states of the machine based on information obtained from the machine, an abnormality level calculation unit that calculates the abnormality levels of the sensor signals for each operating state of the machine determined by the operating state determination unit, and a factor identification unit that determines a factor in an abnormality in the machine from historical data being a series of the abnormality levels for each operating state.