Wire EDM Control Using Machine-Learned Gap Correction Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing control systems for wire electric discharge machines face challenges in maintaining a constant discharge gap, leading to machining precision issues due to the complexity of correlating inter-electrode average voltage and discharge delay time with the discharge gap, requiring extensive experimentation to determine correction parameters for varying variables.
Innovation Solution
A machine learning device is implemented to autonomously learn and optimize correction parameters by observing condition data and determination data, using a multilayer structure to identify correlations and update models, thereby reducing errors and determining optimal correction parameters for machining precision.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If control is performed based on detected values of alternative indicators (inter-electrode average voltage and discharge delay time), then the discharge gap can be maintained constant in linear portions, but the correlation between these indicators and discharge gap collapses at corner portions, leading to machining precision degradation
Solution Approach 1:
The system pre-calculates and stores correction parameters for various machining conditions (wire diameter, workpiece thickness, machining portion type) before actual machining. This allows the control device to quickly retrieve and apply appropriate correction parameters without real-time computation, enabling seamless adaptation between linear and corner portions while maintaining discharge gap consistency.
Solution Approach 2:
The invention applies different correction parameters specifically tailored to different machining portions (linear vs. corner). By identifying the current machining portion type and applying location-specific correction parameters, the system maintains optimal discharge gap control for each specific area, addressing the collapse of correlation at corner portions while preserving effectiveness in linear portions.
2Measurement precision
If extensive experimentation is conducted to determine correction parameters for various variables (programmed shape, wire diameter, workpiece thickness), then accurate correction parameters can be obtained, but the development time and man-hours required become enormous
Solution Approach 1:
Correction parameters for various machining conditions are pre-calculated and stored in a database during the machine setup phase or initial operation. This preliminary preparation eliminates the need for extensive real-time experimentation, as the system can quickly retrieve pre-determined accurate parameters based on the current machining conditions (wire diameter, workpiece thickness, shape type).
Solution Approach 2:
The system creates a database of correction parameters based on experimental data from similar machining conditions. Once parameters are determined for a specific set of conditions, they are stored and reused for identical or similar future machining tasks, eliminating the need to repeat extensive experimentation for each new workpiece and enabling rapid deployment of accurate parameters.
3Reliability
If correction parameters are determined through exhaustive experimentation with widely varying variables, then accurate control can be achieved for known conditions, but the system cannot respond to machining that uses variables not covered by prior experiments
Solution Approach 1:
The correction parameter database is designed to cover a wide range of machining variables (wire diameter, workpiece thickness, machining portion type, shape category). By organizing parameters universally across multiple variables, the system can handle diverse machining conditions with a single unified approach, enabling both reliable control for known conditions and adaptability to new conditions through database lookup.
Solution Approach 2:
The system dynamically selects appropriate correction parameters based on real-time machining conditions (current wire diameter, workpiece thickness, detected machining portion type). This dynamic selection mechanism allows the system to adapt to varying conditions during machining operations, maintaining reliability for known conditions while accommodating new conditions through real-time parameter retrieval and adjustment.
Data Source
AI summary
A control device of a wire electric discharge machine and a machine learning device are provided that can appropriately and readily determine a correction parameter. The control device, which optimizes the correction parameter for wire electrical discharge machining process, includes a machine learning device configured to learn the correction parameter for the wire electrical discharge machining process. The machine learning device includes a state observation unit configured to observe, as a state variable, condition data indicative of a condition for the wire electrical discharge machining process, a determination data acquisition unit configured to acquire determination data indicative of the correction parameter of the case where machining precision is favorable in the wire electrical discharge machining process, and a learning unit configured to learn the correction parameter in association with the condition for the wire electrical discharge machining process using the state variable and the determination data.