Machining Condition Search Using Early Convergence Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional machining condition search techniques require extensive time to find optimal conditions due to the need for machining apparatuses to perform under all conditions until vibrational changes in machining results settle, leading to inefficient optimization processes.
Innovation Solution
A machining condition search device that generates and tests machining conditions, collects results, calculates provisional evaluation values, determines convergence, and terminates or continues machining based on evaluation value stability, allowing for early prediction and optimization of machining conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machining is continuously performed until vibrational change in machining result settles for each machining condition, then evaluation value accuracy is improved, but time consumption increases significantly
Solution Approach 1:
The system performs preliminary machining for a predetermined period to obtain initial evaluation values before the vibrational changes settle. These preliminary values are used to predict the final evaluation values, allowing the system to avoid waiting for complete convergence while still achieving accurate predictions of optimal machining conditions.
Solution Approach 2:
Instead of directly measuring the final stabilized evaluation value which requires long machining time, the system creates a predictive model that copies the relationship between machining parameters and evaluation outcomes. This model is trained on preliminary data and can predict final evaluation values without requiring the actual machining to complete convergence.
2Reliability
If all machining conditions are fully tested until convergence, then optimal condition reliability is improved, but productivity decreases
Solution Approach 1:
The system performs partial machining for a predetermined period rather than waiting for complete convergence. This partial action is sufficient to gather data for prediction, and the system accepts that not all conditions need full convergence testing to identify optimal parameters, thus improving productivity while maintaining reliable predictions.
Solution Approach 2:
The system uses feedback from preliminary evaluation values to iteratively refine predictions of final evaluation values. By continuously updating predictions based on observed trends during the predetermined machining period, the system can reliably identify optimal conditions without requiring all conditions to reach full convergence.
Data Source
AI summary
A machining result processing device includes processing circuitry configured to collect machining result information; calculate a provisional evaluation value for machining performed; estimate an estimated convergence value when the provisional evaluation value has not converged; determine whether to terminate the machining before the provisional evaluation value converges when the provisional evaluation value has not converged; determine the estimated convergence value as an evaluation value when the machining is terminated and determine the convergence value of the provisional evaluation value as an evaluation value after the provisional evaluation value has converged when the machining is not terminated; and determine an optimal machining condition when the search is terminated and generates a machining condition to be tried next when the search is not terminated, in which until it is determined to end the search, each of aforementioned processes described above is repeatedly performed.