Servo Deterioration Diagnosis Using Partial-Period Data Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current deterioration diagnosis methods for servo systems lack accuracy due to the use of all input values over time, which can include interfering features, leading to misleading results.
Innovation Solution
A data processing system that acquires control and detection values from a servo system, extracts diagnostic parameters from specific partial periods of these values, and generates time-series data to improve diagnosis accuracy by filtering out interfering features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If all input values over time are used for deterioration diagnosis, then the quantity of diagnostic data is increased, but the accuracy of diagnosis deteriorates due to interfering features
Solution Approach 1:
The patent segments the time-series input values into multiple periods and selectively extracts values from specific periods that are most indicative of deterioration. This segmentation allows the system to maintain sufficient diagnostic data quantity while eliminating interfering features from irrelevant time periods, thereby resolving the contradiction between data quantity and diagnostic accuracy.
Solution Approach 2:
The patent extracts only the necessary diagnostic values from specific periods of the input time-series data, rather than using all available data. This extraction process removes interfering features while preserving the essential diagnostic information, thus improving accuracy without sacrificing the necessary data quantity for reliable diagnosis.
2Loss of information
If all input values over time are used for deterioration diagnosis, then the completeness of diagnostic information is improved, but the reliability of diagnosis deteriorates due to misleading interfering features
Solution Approach 1:
By segmenting the time-series data into distinct periods and selecting specific periods for extraction, the patent ensures that only relevant diagnostic information is used. This segmentation maintains completeness of meaningful diagnostic information while excluding misleading interfering features from other periods, thereby improving reliability.
Solution Approach 2:
The patent applies local quality by assigning different importance to different time periods, extracting values from periods where deterioration indicators are most prominent. This approach ensures that the diagnostic information from critical periods is preserved while filtering out noise from less relevant periods, thus enhancing both completeness and reliability.
3Measurement precision
If extraction processing is performed on input values to generate diagnostic parameters, then the accuracy of deterioration diagnosis is improved, but the complexity of the processing system increases
Solution Approach 1:
The segmentation of time-series data into periods with clear physical or operational meanings simplifies the extraction process. By focusing on specific periods rather than processing the entire time-series, the system reduces computational complexity while maintaining high diagnostic accuracy through targeted extraction of relevant features.
Solution Approach 2:
The patent applies partial action by extracting only the necessary portions of the input data from specific periods rather than processing all data. This selective approach achieves high diagnostic accuracy without requiring complex processing of the entire dataset, thus balancing accuracy with system simplicity.
Data Source
AI summary
Accuracy of deterioration diagnosis is improved. Data processing system (1) of servo system (7) includes acquisition part (2) and generator (3). Acquisition part (2) acquires, as input values, at least one of a control value and a detection value. Generator (3) generates a diagnostic parameter by performing at least extraction processing on the input value acquired by acquisition part (2). The diagnostic parameter is used for deterioration diagnosis of object (OB1) of at least one of load (73) and servomotor (72) in servo system (7). The input values are values that vary with time. In the extraction processing, generator (3) extracts, as a diagnostic parameter, a value in a partial period among the input values.