Machining Apparatus Machine Learning Optimization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional machining processes require significant operator effort to adjust machining conditions for optimal results, as these conditions vary with tool characteristics, workpiece characteristics, and machining types, leading to inefficiencies even with the reuse of past machining conditions.

Innovation Solution

Integration of a machine learning device within the machining apparatus that uses reinforcement learning to adjust machining conditions based on measured machining time and accuracy, providing rewards for optimal performance and allowing for the sharing of learning results between machines.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If operator adjusts machining conditions by trial and error using past machining conditions from database, then machining conditions can be reused to reduce adjustment efforts, but operator still needs to repeat trial and error to adjust past conditions to suit current machining situation

Engineering Contradiction:

Improveadjustment of machining conditionsVSAvoidtime for trial and error adjustment

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The machine learning device automatically adjusts machining conditions by learning from past machining data and current machining situation, eliminating the need for operator trial and error. The system serves itself by autonomously optimizing parameters based on accumulated knowledge and real-time feedback.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system implements feedback by measuring actual machining results and using this information to continuously improve future machining condition adjustments. The machine learning device learns from the outcomes of previous machining operations to automatically optimize conditions without requiring operator intervention.

Inventive Principle:

Principle #23Feedback

2Extent of automation

If machine learning device automatically adjusts machining conditions based on measured machining time and accuracy, then optimal machining conditions can be calculated automatically, but system complexity increases

Engineering Contradiction:

Improveadjustment of machining conditionsVSAvoidmachining apparatus system

Core Design Contradiction:

Extent of automationVSDevice complexity

Solution Approach 1:

The machine learning device serves multiple functions: it measures machining time, evaluates machining accuracy, stores past machining data, and automatically adjusts machining conditions. By consolidating these functions into a single integrated system, the patent reduces overall system complexity despite the advanced capabilities provided.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS10180667B2Controller-equipped machining apparatus having machining time measurement function and on-machine measurement function

Publication Date: 2019.01.15 FANUC LTD

AI summary

A machining apparatus is provided with a machine learning device that performs machine learning. The machine learning device performs the machine learning by receiving the input of machining accuracy between a machining shape of a workpiece measured on-machine and design data on the workpiece and machining time of the workpiece measured by a measurement device. Based on a result of the machine learning, the machining apparatus changes machining conditions such that the machining accuracy increases and the machining time becomes as short as possible.