Machine Learning Device for Chatter Detection in CNC Tools
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing chatter detection methods in machine tools lack sufficient accuracy in detecting chatter occurrence modes that vary with machining programs and tool usage status, requiring inappropriate reference value settings and failing to flexibly respond to different chatter scenarios.
Innovation Solution
A machine learning device that observes state variables such as tool vibration, building vibration, audible sound, acoustic emission, and motor control current, using unsupervised learning to generate a learning model and output scores for normal or abnormal conditions, enabling accurate chatter detection without requiring specific reference value settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional chatter detection methods using fixed threshold values are used, then the detection process is simple, but the detection accuracy is insufficient when chatter occurrence modes vary with machining programs and tool usage status
Solution Approach 1:
The system performs self-learning by automatically observing vibration patterns during normal machining operations and building its own detection model without requiring manual intervention or pre-set thresholds. The learning unit accumulates vibration data and autonomously determines what constitutes normal versus abnormal vibration patterns, enabling the system to adapt to different machining conditions automatically.
Solution Approach 2:
The system transitions from using fixed threshold parameters to dynamic parameters that change based on observed vibration patterns. By continuously learning from operational data, the detection thresholds and criteria are automatically adjusted to match the specific machining program and tool usage status, thereby improving detection accuracy across varying conditions.
2Measurement precision
If multiple state variables are observed to improve detection accuracy, then chatter detection becomes more accurate, but the system complexity increases
Solution Approach 1:
The system merges multiple state variable observations (tool vibration, machine tool vibration, building vibration, audible sound, acoustic emission, and motor control current) into a unified learning model. By combining these diverse data sources, the system achieves comprehensive chatter detection that leverages the complementary information from each sensor type, improving overall detection accuracy while managing complexity through integrated processing.
3Adaptability or versatility
If fixed reference values are used for chatter detection, then the system is easy to operate, but it cannot flexibly respond to different chatter occurrence modes
Solution Approach 1:
The system performs preliminary learning during normal machining operations to establish a baseline understanding of vibration patterns before actual chatter detection is needed. By pre-learning what constitutes normal operation across different machining programs and tool conditions, the system is prepared to flexibly respond to chatter occurrences without requiring manual reconfiguration or complex user intervention during operation.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution effectively reduces the number of poor workpieces by accurately detecting chatter occurrence, including tool damage, with high accuracy and flexibility across various machining conditions.
Implementation Method 1
a vibration sensor (11) for detecting the vibration of a tool (1a)
Implementation Method 2
an audible sound sensor (12) for detecting an audible sound
Implementation Method 3
an acoustic emission sensor (13) for detecting an acoustic emission
Data Source
AI summary
A machine learning device for detecting an indication of an occurrence of chatter in a tool for a machine tool, includes a state observation unit which observes at least one state variable of a vibration of the machine tool itself, a vibration of a building in which the machine tool is installed, an audible sound, an acoustic emission and a motor control current value of the machine tool, in addition to a vibration of the tool; and a learning unit which generates a learning model based on the state variable observed by the state observation unit.