Machine Learning for Coil Winding Parameter Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional coil producing apparatuses require manual setting of detailed operating conditions by operators, which is labor-intensive and time-consuming, leading to inefficiencies in coil winding.
Innovation Solution
A machine learning apparatus that communicates with a winding machine, utilizing a state observing unit and a learning unit to automatically determine optimal winding parameters such as dimension, resistance, speed, and tension values through reinforcement learning, eliminating the need for manual input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual setting of operating conditions is used, then the coil winding process can be controlled, but labor intensity and time consumption increase significantly
Solution Approach 1:
The system performs self-learning by automatically observing state variables and command values, then autonomously determines optimal operating conditions without requiring manual intervention. The learning unit stores and updates determination results, enabling the system to self-adjust and eliminate the need for repeated manual setup while reducing operation time and labor intensity
Solution Approach 2:
The patent replaces manual mechanical adjustment with an automated learning system that uses observation and determination units to automatically set operating conditions. This substitution of human operation with an intelligent system reduces both time consumption and labor intensity while maintaining precise control
2Manufacturing precision
If manual trial and error method is used to set operating conditions, then the winding parameters can be adjusted, but production efficiency decreases due to labor intensity
Solution Approach 1:
The system implements feedback by observing actual state variables (dimension, resistance, wire rod used amount, program execution time) and comparing them with command values. The learning unit uses this feedback information to determine and store optimal operating conditions, replacing trial-and-error methods with a systematic feedback-driven approach that improves both precision and productivity
Solution Approach 2:
The system performs preliminary learning and determination of optimal operating conditions before actual production. By pre-establishing determination results through observation and analysis, the system eliminates the need for time-consuming trial and error during production, thereby improving manufacturing precision while increasing overall productivity
3Reliability
If detailed operating conditions are manually configured, then coil winding control is achieved, but many man-hours are required
Solution Approach 1:
The system automatically observes state variables and command values, then self-determines optimal operating conditions without human intervention. The learning unit stores these determination results for future use, enabling the system to maintain reliable coil winding control while eliminating the need for manual configuration and reducing man-hours required
Solution Approach 2:
The system creates a model of optimal operating conditions by copying and storing the relationship between observed state variables and effective command values. This copied knowledge is stored in the learning unit and reused to control coil winding reliably without requiring repeated manual setup, thereby reducing time loss while maintaining control reliability
Data Source
AI summary
A machine learning apparatus includes a state observing unit for observing a state variable comprised of at least one of an actual dimension value, a resistance actual value, etc., and at least one of a dimension command value, a resistance command value, etc., and an execution time command value for a program, and a learning unit for performing a learning operation by linking at least one of an actual dimension value, a resistance actual value, etc., to at least one of a dimension command value, a resistance command value, etc., observed by the state observing unit, and an execution time command value for the program.