Machine Learning Control of CNC Axis Acceleration for Machining Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machining technologies face challenges in quantitatively controlling target tolerance and machined surface quality due to the need for speed distribution thresholds, which are not easily set, and require resetting when machine tools or machining purposes change.
Innovation Solution
An acceleration/deceleration adjustment device using machine learning to optimize parameter combinations for shaft movement control, including N-order time differential elements, with a state observer, determination condition acquirer, reward calculation unit, value function update unit, and decision maker to adjust parameters based on machining accuracy and time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If speed distribution thresholds are used as a criterion for parameter adjustment, then parameter optimization can be achieved, but it becomes difficult to quantitatively control target tolerance and machined surface quality
Solution Approach 1:
The patent changes the control parameters from speed distribution thresholds to direct machining accuracy parameters (shape error, positional deviation). This allows operators to directly specify desired machining accuracy levels without needing to understand complex speed distribution settings, thereby improving ease of operation while maintaining manufacturing precision control
Solution Approach 2:
The patent introduces a machine learning model as an intermediary that translates desired machining accuracy requirements into optimal parameter settings. The model acts as a mediator between the operator's accuracy requirements and the complex control parameters, eliminating the need for operators to directly set speed distribution thresholds
2Manufacturing precision
If appropriate speed distribution is set for a predetermined machining purpose, then machining quality can be optimized, but it requires resetting parameters when machine tool or machining purpose changes
Solution Approach 1:
The patent creates a universal parameter setting system based on machining accuracy requirements that can be applied across different machine tools and machining purposes. Instead of maintaining separate parameter sets for each application, the system uses a unified approach where the machine learning model adapts to different scenarios while operators simply specify their desired accuracy levels
Solution Approach 2:
The patent implements dynamic parameter adjustment through machine learning that automatically adapts to different machine tools and machining purposes in real-time. The system continuously learns from operational data and adjusts parameters dynamically based on the specific context, eliminating the need for manual resetting when conditions change
3Manufacturing precision
If operator manually adjusts acceleration/deceleration time constant and speed parameters, then machining error and surface quality can be checked, but the process is time-consuming and lacks quantitative control
Solution Approach 1:
The patent implements a self-service system where the machine learning model automatically performs parameter optimization based on desired accuracy requirements. The system self-adjusts parameters without requiring manual trial-and-error by operators, thereby reducing adjustment time while maintaining or improving machining precision control
Solution Approach 2:
The patent establishes a feedback loop where the machine learning model continuously monitors machining results and automatically adjusts parameters to achieve target accuracy levels. This closed-loop system eliminates time-consuming manual adjustments by using real-time feedback to drive automatic parameter optimization
Data Source
AI summary
A machine learning device that estimates parameters that relate to the control of the movement amount in each control cycle and include Nth-order time differential elements of each axis of a machine tool that machines workpieces, the machine learning device comprising: a state observation unit that observes data indicating the movement state of the machine tool; a determination condition acquisition unit that acquires, as determination data, target values relating to the data observed by the state observation unit; a reward calculation unit that calculates, on the basis of the state data and the determination data, rewards associated with machining based on the parameters; a value function update unit that updates value functions on the basis of the rewards; and a decision-making unit that estimates, on the basis of the value functions, combinations of set values of the parameters that are more suitable for the machining, and outputs said combinations.