Signal Correlation Diagnosis for Faster, More Accurate Fault Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing abnormalities in processing systems using time-series data are insufficient in eliminating erroneous detection and improving accuracy.
Innovation Solution
A diagnostic device that acquires input signals and diagnoses abnormalities using a combination of first and second index values, where the first index value indicates the similarity of a target signal to a reference waveform and the second index value is based on the comparison of the target signal with a predetermined pattern, calculated using values of multiple input signals, with weighting coefficients adjusted according to the correlation between signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If a single time-series data item is used for diagnosis, then quick detection of minor troubles is achieved, but accuracy of diagnosing abnormality is insufficient
Solution Approach 1:
The patent segments the diagnosis process into two independent index calculations: a first index based on individual time-series data comparison with reference data, and a second index based on multi-dimensional pattern matching across multiple data items. This segmentation allows each index to specialize in different aspects of anomaly detection, resolving the contradiction between speed and accuracy by combining their results.
Solution Approach 2:
The patent merges the results of two different diagnostic approaches (first index from individual data comparison and second index from collective pattern analysis) into a unified diagnosis. By combining these complementary methods, the system achieves both the quick detection capability of single-data analysis and the high accuracy of multi-data analysis.
2Measurement precision
If multiple time-series data items are analyzed collectively, then detection of overall abnormal states is achieved, but quick detection of minor troubles is reduced
Solution Approach 1:
The patent segments the diagnostic analysis into two parallel pathways: one that quickly compares individual data items against reference data (first index), and another that performs more comprehensive multi-dimensional pattern matching (second index). This segmentation enables the system to maintain detection speed through the first pathway while achieving high accuracy through the second pathway.
Solution Approach 2:
The patent applies partial action by using the first index for quick screening of individual data items and the second index only when needed for comprehensive pattern recognition. This selective application of analysis depth maintains overall system speed while achieving accurate diagnosis when necessary.
3Speed
If only first index value (individual signal similarity) is used, then quick detection is achieved, but erroneous detection increases
Solution Approach 1:
The patent implements feedback by using the second index (based on multi-dimensional pattern correlation) to verify and correct diagnoses made by the first index. When the first index indicates a potential anomaly, the second index provides feedback to confirm whether it represents a true abnormality or an error, thereby reducing false detection rates.
Solution Approach 2:
The patent merges the first index (individual data comparison) and second index (collective pattern analysis) into a unified diagnostic decision. This combination allows the system to benefit from the speed of individual comparison while using collective pattern recognition to filter out erroneous detections, improving overall reliability.
4Measurement precision
If only second index value (collective pattern comparison) is used, then accurate detection of overall abnormalities is achieved, but detection of individual minor troubles is reduced
Solution Approach 1:
The patent segments the detection capability into two specialized components: the first index that excels at detecting individual minor troubles through direct comparison with reference data, and the second index that excels at detecting overall abnormal states through multi-dimensional pattern analysis. This segmentation ensures that neither detection capability is lost.
Solution Approach 2:
The patent merges the strengths of both index types into a unified diagnostic system. The first index provides sensitivity to individual minor changes, while the second index provides robustness for detecting overall abnormalities. Together, they overcome the limitations of using either method alone.
Data Source
AI summary
A diagnostic device (10) includes an acquirer (101) and a diagnoser (140). The acquirer (101) acquires a plurality of input signals including a target signal to be diagnosed for abnormality. The diagnoser (140) diagnoses, using a first index value relating to the target signal and a second index value relating to the plurality of input signals based on a correlation between the plurality of input signals, whether an abnormality occurs. The first index value indicates a degree of similarity of a waveform of the target signal to a predetermined reference waveform. The second index value is a value that is based on comparison between the target signal and a predetermined pattern and is calculated from values of the plurality of input signals.