Machine Learning Control for Burr-Minimizing Cutting Conditions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Burr formation on workpieces during cutting processes leads to tool wear and increased processing time, necessitating a reduction in burr occurrence.
Innovation Solution
A machine learning apparatus that generates a learning model to estimate optimal cutting conditions by acquiring information on workpiece shape, material, cutting path, tool type, and tool wear, and evaluates burr formation to output conditions that minimize burr occurrence, using algorithms like convolutional neural networks or Q-learning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If a tool is used to eliminate burrs on the workpiece, then the burrs are removed, but the tool becomes worn out and the elimination time increases
Solution Approach 1:
The learning model predicts optimal cutting conditions before the cutting process occurs, preventing burr formation in the first place rather than requiring subsequent burr elimination operations. This preliminary prediction and prevention approach avoids the need for tool-based burr removal, thereby eliminating tool wear and reducing total processing time
2Productivity
If more burrs occur on the workpiece, then the cutting process can be performed, but the time needed for burr elimination becomes longer
Solution Approach 1:
The learning model is trained using feedback from historical cutting data including workpiece information, cutting conditions, and burr occurrence evaluations. This feedback loop enables the model to continuously improve its predictions of optimal cutting conditions that minimize burr formation, thereby reducing the time needed for burr elimination and improving overall productivity
3Reliability
If the tool is used repeatedly to eliminate burrs, then burrs are removed, but the tool wear increases and elimination time increases
Solution Approach 1:
By predicting optimal cutting conditions before the cutting process using the learning model, the system prevents burr formation at the source. This preliminary action eliminates the need for repeated tool-based burr removal operations, maintaining workpiece quality while reducing total processing time and tool wear
Data Source
AI summary
A machine learning apparatus includes a first information acquiring unit that acquires first information including at least one of a shape of a workpiece, a material of the workpiece, a cutting path of a cutting process, a type of a tool, and an amount of wear of the tool; a second information acquiring unit that acquires second information correlated with an evaluation of a burr occurring on the workpiece due to the cutting process; and a learning unit that executes learning processing using a plurality of pieces of the first information and a plurality of pieces of the second information, and generates a learning model that outputs a cutting condition, according to another piece of first information that is different from the plurality of pieces of first information.