Signal Portion Comparison for Non-Destructive Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing non-destructive testing methods are cumbersome, operator-dependent, and lack versatility in detecting defects across various components, particularly when dealing with large numbers of similar elements, as they often require extensive development and are not effective for unanticipated defects.
Innovation Solution
A method utilizing dynamic time warping (DTW) to calculate distances between signal portions from multiple elements, followed by statistical analysis of these distances to identify anomalies, employing measures of central tendency and dispersion, and comparing these indicators across elements to determine soundness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional non-destructive testing methods are used with human operators analyzing measurement signals, then detection can be performed, but the inspection becomes very lengthy and tedious with quality depending on operator skill
Solution Approach 1:
The system performs self-diagnosis by automatically comparing measurement signals against reference signals and statistical models to detect anomalies, eliminating the need for human operators to manually analyze each signal while maintaining high detection quality through algorithmic consistency
Solution Approach 2:
The patent replaces the mechanical human operator analysis process with an automated computational system that uses signal processing algorithms, statistical analysis, and anomaly detection methods to evaluate measurement signals, thereby eliminating operator-dependent variability and reducing inspection time
2Measurement precision
If existing non-destructive testing methods are developed for specific applications, then detection accuracy for those applications can be improved, but the methods are not suitable for detecting unanticipated defects or inspecting other components
Solution Approach 1:
The patent creates a universal anomaly detection framework that can be applied to any measurement signal from any component type by using generic signal processing and statistical comparison methods, allowing the same system to detect both anticipated and unanticipated defects across diverse applications without requiring application-specific customization
Solution Approach 2:
The system adapts to different applications by dynamically adjusting statistical parameters such as mean, standard deviation, and anomaly thresholds based on the specific measurement signal characteristics being analyzed, enabling accurate detection across varying component types and defect scenarios
3Ease of manufacture
If statistical values are calculated for each time sample to construct a temporally homogenized signal, then detection can be performed, but the method is not effective when applied to inspection of other components where defects generate variations in signal shape rather than amplitude increases
Solution Approach 1:
The patent employs dynamic time warping to create a time-varying alignment between reference and measurement signals, allowing the comparison to adapt to temporal distortions and shape variations caused by different defect types, thereby maintaining detection effectiveness across diverse component geometries and defect characteristics
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A non-destructive testing method for a plurality of elements based on measurement signals, each measurement signal corresponding to a different determined element, comprising the following steps: - obtaining (S01) a sequence of at least one signal portion from each measurement signal, - for each pair associating two signal portions having the same position in their respective sequence, a distance between these signal portions of said pair is calculated (S02), said distance being calculated using a dynamic time deformation, - for each signal portion, determining (S03) at least one statistical indicator from the distances of pairs involving said signal portion, the statistical indicator being associated with the element of said signal portion, - for each element, comparing (S04) at least one indicator of said element with indicators of other elements, and based on the comparison,determination (S05) of whether the element is healthy or unhealthy.