Machining Diagnosis Using Cycle-Based Characteristic Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for diagnosing abnormalities in machining devices are limited in providing versatile evaluation, as they primarily display chatter estimation information and detected chatter spectra without comprehensive analysis.
Innovation Solution
A diagnosis system that includes a diagnosis apparatus connected to a machining device, which extracts characteristic information from detected physical quantities, generates models for normal operation, and determines abnormality by using context and detection information, enabling comprehensive evaluation and display of time-series data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If only chatter estimation information and detected chatter spectrum are displayed, then the display is simple, but the evaluation is not versatile
Solution Approach 1:
The diagnosis apparatus performs multiple functions including detecting physical quantities, extracting characteristic information, generating models, determining abnormalities, and displaying results. This multi-functional approach enables versatile evaluation while integrating all functions into a unified system that manages complexity through functional integration
2Measurement precision
If comprehensive analysis and model generation are performed, then evaluation versatility is improved, but processing time increases
Solution Approach 1:
The system generates models during normal operation based on detected physical quantities and characteristic information. This preliminary model generation enables rapid abnormality determination when needed, as the comparison framework is already established, reducing processing time during actual diagnosis while maintaining comprehensive analysis capabilities
Solution Approach 2:
The system continuously monitors physical quantities, compares them against generated models, and provides feedback on abnormality determinations. This feedback mechanism allows the system to learn and refine its models over time, improving diagnosis precision while optimizing processing efficiency through iterative refinement rather than repeated comprehensive analysis
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An information processing apparatus (100) includes: a detected information receiver (112) configured to obtain physical quantity information indicating a physical quantity that occurs in a target device (200) and changes in accordance with operation of the target device; a characteristic information generator (102a) configured to generate characteristic information indicating a characteristic of the physical quantity information; a data manager (103) configured to classify the characteristic information into a plurality of stages contained in one cycle, so as to generate a plurality of pieces of characteristic information for each cycle, the cycle being a cycle including the plurality of stages and to be repeatedly executed by the target device; a storage unit (113) configured to store the plurality of pieces of characteristic information for each cycle; and an output control unit (104) configured to output one or more pieces of the plurality of pieces of characteristic information for a plurality of cycles that are different to an output unit (115).