Machine Tool Wear Estimation From Time-Series Pattern Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to reliably determine machine properties without direct metrological capture, particularly in dynamic manufacturing environments where wear and tear lead to unexpected failures and quality deficits.
Innovation Solution
A computer-implemented method that captures time series of physical measurement variables, detects change points, extracts pattern-sequence instances, and generates classes to identify machine properties without direct metrological measurement, enabling the determination of wear levels and optimizing machine operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If statistical calculations are used to determine machine properties, then wear levels can be estimated, but the reliability decreases when changing operating modes or manufacturing different components
Solution Approach 1:
The method segments the time series data into pattern-sequence instances using change points as delimiters. This segmentation allows the system to analyze specific operational patterns independently, making the wear determination adaptable to different operating modes while maintaining reliability through pattern-specific analysis.
Solution Approach 2:
The system changes parameters by transforming raw time series data into pattern-sequence classes through multiple processing steps (change point detection, pattern extraction, class formation). This parameter transformation enables the same analytical framework to handle diverse operating conditions by adapting to the statistical characteristics of different data patterns.
2Measurement precision
If direct metrological capture of machine properties is implemented, then measurement precision improves, but device complexity and cost increase
Solution Approach 1:
The method introduces an intermediary approach by using pattern-sequence classes as mediators between raw sensor data and machine property determination. Instead of directly measuring wear with complex metrological equipment, the system uses intermediate pattern recognition and statistical analysis to infer wear levels from readily available operational data.
Solution Approach 2:
The system replaces direct mechanical/metrological measurement systems with an information-processing approach. By substituting physical measurement equipment with computational analysis of time series data, the method achieves wear determination without the complexity and cost of specialized measurement devices.
3Productivity
If existing methods (US 2016/0 091 393 A1) are used to analyze operating data, then machining methods can be identified, but the ability to dynamically calculate wear levels across different components is limited
Solution Approach 1:
The method achieves universality by creating a multi-functional analytical framework that can process time series data from various machining operations through the same pipeline (change point detection, pattern extraction, class formation). This universal approach enables dynamic wear calculation across different components and operating modes while maintaining high detection efficiency.
Data Source
AI summary
A computer-implemented method determines a property of a machine, in particular a machine tool, without metrologically capturing the property. The method includes the following steps:capturing one or more first time series of one or more physical measurement variables of the machine;detecting change points in the one or more first time series;extracting pattern-sequence instances from the first time series on the basis of the detected change points;producing a plurality of pattern-sequence classes in accordance with the extracted pattern-sequence instances;identifying at least one characteristic of a plurality of pattern-sequence instances of the same pattern-sequence class and a time curve of the characteristic;determining a property of a machine using the determined characteristic and/or using the time curve of the determined characteristic.