Damage Analysis Data Segmentation Using Time-Window Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information processing devices struggle to efficiently analyze the degree of damage to vehicle components using large volumes of sensor data, requiring extensive computation time and lacking effective methods to extract relevant data for rapid analysis.
Innovation Solution
The device employs a method to segment original data into multiple datasets using time windows, calculating frequency distributions of specific features, and determining similarity with extracted data to analyze damage more efficiently, utilizing processing circuitry to divide and segment data based on defined features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If original data collected over a specified period is used for damage analysis, then analysis accuracy is improved, but analysis time increases
Solution Approach 1:
The original data collected over a specified period is divided into multiple datasets using time windows. The processing circuitry segments the continuous data stream into discrete time-based portions, allowing selective analysis of representative segments rather than processing all original data, thus reducing analysis time while maintaining damage assessment accuracy through frequency distribution comparison.
Solution Approach 2:
The patent extracts specific features from the original data, particularly the first feature (physical quantity related to damage) and second feature (other physical quantity). By extracting and analyzing only the relevant features rather than the complete original dataset, the system achieves efficient damage analysis with reduced computation time while preserving analytical accuracy.
2Productivity
If data is segmented using time windows, then processing speed is improved, but data similarity to original data must be maintained
Solution Approach 1:
The processing circuitry calculates frequency distributions for both the extracted data from time windows and the original data, then compares these distributions to determine similarity. This feedback mechanism ensures that the segmented data maintains representativeness of the original data characteristics, validating that the time window segmentation preserves the essential patterns needed for accurate damage analysis.
Data Source
AI summary
The information processing device includes a step of setting a plurality of time windows for segmenting data from the original data using two physical quantities related to damage to the device or the component as a first feature and a second feature, a step of segmenting data from the original data, a step of calculating a frequency distribution in the original data and a frequency distribution in the extracted data for the first feature divided into a plurality of parts by the second feature, and a step of determining whether the original data and the extracted data are similar using the frequency distributions. The information processing apparatus repeatedly executes these steps while changing the setting of a plurality of time windows, and outputs extracted data similar to the original data.