Motor Damage Data Extraction Using Time-Window Frequency Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information processing apparatuses fail to capture features of entire original data beyond vehicle speed, necessitating a solution that can extract data with features of entire original data including multiple feature amounts.
Innovation Solution
An information processing apparatus that acquires data using multiple sensors, calculates an indicator value for motor damage, and performs search processing with time windows to extract data with errors less than a threshold, allowing clustering and relative frequency distribution calculations to reduce data amount while maintaining accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is compressed by extracting only vehicle speed data, then data amount is reduced, but features of entire original data including other feature amounts are lost
Solution Approach 1:
The original data is divided into multiple sections based on time periods, and feature amounts are calculated for each section. This segmentation allows the system to process and represent the entire original data through aggregated section-level features, reducing data amount while preserving comprehensive information about all feature amounts including voltage, current, and other motor parameters.
Solution Approach 2:
The system transforms raw data into derived parameters such as average values, maximum values, minimum values, and standard deviations for each feature amount across different time sections. This parameter transformation compresses the data while maintaining the essential characteristics and features of the original multi-dimensional data including all sensor measurements.
2Measurement precision
If entire original data is used to calculate indicator values, then accuracy is maintained, but processing time increases
Solution Approach 1:
By dividing the analysis period into multiple sections and calculating feature amounts for each section, the system can use aggregated section data instead of processing every individual data point from the original continuous data stream. This segmentation maintains accuracy by preserving key statistical features while dramatically reducing processing time through data compression.
Solution Approach 2:
The system extracts essential feature amounts (average, maximum, minimum, standard deviation) from each time section and uses only these extracted features for indicator value calculation. This extraction process eliminates redundant data while retaining the critical information needed for accurate motor damage assessment.
3Productivity
If data is compressed to reduce amount, then processing speed increases, but accuracy of indicator value calculation decreases
Solution Approach 1:
The system applies multiple parameter transformations (average, maximum, minimum, standard deviation) to each feature amount across time sections. These parameter changes create a compressed representation that captures the essential variability and characteristics of the original data, enabling fast processing while maintaining calculation accuracy for indicator values.
Solution Approach 2:
The calculated feature amounts serve multiple functions: they represent the original data for compression, provide input for indicator value calculation, and enable both fast processing and accurate results. This multi-functionality of the compressed feature data resolves the contradiction between processing speed and accuracy.
Data Source
AI summary
An processing device of an information processing apparatus includes: a first step of calculating a relative frequency distribution of original data; a second step of setting a plurality of time windows for cutting out data of a partial period of the original data; a third step of cutting out data from the original data; a fourth step of calculating a relative frequency distribution in extracted data; and a fifth step of calculating an error between the relative frequency distribution in the original data and the relative frequency distribution in the extracted data, and performs a search process of repeatedly executing a trial from the second to fifth steps by changing the setting of the time windows. The processing device calculates an index value of damage of a drive motor by using the extracted data in which the error becomes equal to or smaller than a threshold value.