Sensor State Classification for Faster Failure Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for detecting failures in devices like elevators and air conditioners face challenges in accurately distinguishing between device abnormalities and changes in operation states, particularly when using principal component analysis, as they fail to efficiently determine the cause of deviations in principal component scores.
Innovation Solution
An operation state classification apparatus that calculates principal components and physical quantities from sensor data, using these to generate effective classifications for failure detection by selecting relevant sensor items and shaping sensor data for accurate operation state classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of sensor items used for classification is increased, then detection accuracy may be improved, but the number of classifications becomes excessive and the same operation state does not occur
Solution Approach 1:
The patent extracts only the essential sensor items needed for operation state classification by selecting sensor items with high contribution to classification accuracy. This reduces the number of sensor items from a large set to a minimal effective subset, preventing excessive classifications while maintaining detection accuracy.
Solution Approach 2:
The patent changes the parameter of sensor item selection by using contribution rate analysis to identify and select only those sensor items that significantly contribute to operation state classification. This parameter-based selection optimizes the balance between detection accuracy and classification manageability.
2Device complexity
If principal component analysis is used for dimension reduction, then the number of classifications does not become excessive, but it cannot be immediately determined whether deviation is due to abnormality or operation state change
Solution Approach 1:
The patent performs preliminary classification of operation states using selected sensor items before conducting failure determination. By pre-establishing operation state classifications, the system can quickly identify whether a deviation corresponds to a known operation state change or indicates an abnormality, eliminating the need for time-consuming re-analysis.
Solution Approach 2:
The patent introduces operation state classification as an intermediary step between principal component analysis and failure determination. This intermediary classification provides context for interpreting principal component scores, enabling rapid distinction between operation state changes and actual failures.
3Measurement precision
If classification is performed for each operation state, then failure detection accuracy is improved, but it becomes difficult to determine whether measurement value exceeding is due to failure or operation state change
Solution Approach 1:
The patent segments the analysis process into distinct stages: operation state classification based on selected sensor items, followed by failure determination within each classified state. This segmentation allows the system to first identify the operation state context, then focus failure detection efforts within that specific context, making cause determination straightforward.
Data Source
AI summary
A principal component calculating unit (13) for calculating a principal component of a plurality of sensor data collected from a device, and a physical quantity for failure determination calculating unit (11) for calculating a physical quantity to be used for failure determination on the basis of the plurality of sensor data are provided. Further, an operation state classification unit (14) for generating classification, for each operation state, of the sensor data collected from the device, using the principal component calculated by the principal component calculating unit (13) and the physical quantity calculated by the physical quantity for failure determination calculating unit (11) is provided.