Numerical Controller Machine Learning Override Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing PID control methods for machining require extensive experimental gain setting adjustments based on machine configuration, tool type, workpiece material, and cutting amount, necessitating repeated attempts to achieve secure control, which is time-consuming and inefficient.
Innovation Solution
A numerical controller with a machine learning device that employs reinforcement learning to automatically adjust override control setting values, using state observation, reward calculation, and machine learning to determine optimal control settings, eliminating the need for manual gain adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If PID control with manual gain adjustment is used, then control stability can be achieved, but extensive experimental gain setting and repeated attempts are required, increasing adjustment time
Solution Approach 1:
The system performs self-learning of optimal gain values through reinforcement learning. The machine learning device automatically adjusts and learns the optimal gain settings for PID control without requiring manual experimental adjustment, enabling the system to serve itself in optimizing control parameters while maintaining stability
Solution Approach 2:
The invention changes the approach from manual parameter adjustment to automated machine learning-based parameter optimization. The system learns optimal gain values (Kp, Ki, Kd) through reinforcement learning algorithms, dynamically determining control parameters without human intervention, thus reducing adjustment time while maintaining reliability
2Reliability
If extensive experimental gain setting is performed, then secure control can be achieved, but the process becomes time-consuming and inefficient
Solution Approach 1:
The invention replaces the mechanical trial-and-error adjustment process with an automated machine learning system. Instead of manually adjusting gains through repeated experiments, the reinforcement learning device automatically determines optimal control parameters, substituting human operational efficiency with automated intelligent systems
Solution Approach 2:
The system implements continuous feedback through the reinforcement learning mechanism. The machine learning device monitors control performance and uses this feedback to iteratively improve gain settings, automatically learning from past performance to optimize future control actions without requiring manual experimental cycles
Data Source
AI summary
A numerical controller has a machine learning device that performs machine learning of the adjustment of a setting value used in override control. The machine learning device acquires state data showing states of the numerical controller and a machine, sets reward conditions, calculates a reward based on the state data and the reward conditions, performs the machine learning of the adjustment of the setting value used in override control, and determines the adjustment of the setting value used in override control, based on a machine learning result and the state data.