Machining Condition Search Using State-Weighted Evaluation Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machining condition searching devices require a large amount of training data and the construction of a learning model to predict the influence of changing machining states on machining results, making it inefficient to search for optimal machining conditions.
Innovation Solution
A machining condition searching device that generates and adjusts machining conditions based on real-time machining states, collects and evaluates machining results, and constructs an evaluation value prediction model to predict optimal machining conditions without the need for extensive data collection or learning model construction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a learning model is constructed to predict machining results based on machining state changes, then prediction accuracy is improved, but a large amount of training data collection and model construction time are required
Solution Approach 1:
The system pre-calculates and stores evaluation values for multiple candidate machining conditions in an evaluation value storage unit. When searching for optimal machining conditions, these pre-calculated values are directly retrieved and compared, eliminating the need for real-time prediction model construction and extensive data collection during the search process
Solution Approach 2:
Instead of constructing a complex learning model from scratch, the system creates a simplified copy by storing pre-calculated evaluation values for different machining conditions. This copy (evaluation value storage) can be quickly queried and compared without requiring extensive training data or model construction time
2Ease of operation
If the same machining condition is used despite machining state changes, then operational simplicity is maintained, but machining result quality deteriorates
Solution Approach 1:
The machining condition searching device automatically monitors machining state changes and autonomously determines optimal machining conditions by comparing pre-stored evaluation values. This self-service mechanism eliminates the need for operator intervention while maintaining machining quality, allowing simple operation without sacrificing precision
Solution Approach 2:
The system continuously monitors machining state changes and uses this feedback to select appropriate machining conditions from pre-calculated evaluation values. This feedback loop ensures machining quality is maintained while keeping the operation simple for the user
Data Source
AI summary
A machining condition searching method includes: generating a machining condition to be set in a machining device; collecting a machining state; collecting the machining result performed under the machining condition; calculating an evaluation value of the machining on the basis of the machining result; constructing an evaluation value prediction model predicting, on the basis of the machining condition, the machining state, and the evaluation value, an evaluation value corresponding to the machining condition that has not been tried; and constructing the evaluation value prediction model on the basis of a change degree in the relationship between the machining condition and the evaluation value, and performs weighting based on the machining state on the evaluation value prediction model. The machining condition to be tried next is generated using a predictive value of the evaluation value. Each of the above processes is repeatedly performed until it is determined to end a search.