Machine Learning for Laminated Core Stacking Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional laminated core manufacturing methods using robots result in stacking errors and require multiple mold jigs, leading to inconsistent quality and dimensional changes over time, making it difficult to maintain high-quality core production.
Innovation Solution
A machine learning device that observes states of core sheets and manufacturing apparatuses, updates manipulated variables for stacking, and uses reinforcement learning to optimize the stacking process, incorporating a robot vision system for dimension detection and error correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If multiple mold jigs are used for different types of laminated cores, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent implements a single mold jig that can accommodate multiple types of laminated cores by using adjustable positioning mechanisms and programmable robot control. The robot control unit stores multiple moving paths for different core types, allowing one mold jig to perform the function of multiple specialized jigs, thereby reducing device complexity while maintaining manufacturing precision.
Solution Approach 2:
The mold jig incorporates adjustable and reconfigurable components that can be dynamically adjusted between different core types. The robot system dynamically selects appropriate moving paths and positioning parameters based on the specific core type being manufactured, enabling the mold jig to adapt to different manufacturing requirements without requiring multiple fixed configurations.
2Reliability
If mold jigs are used over extended periods, then manufacturing stability is improved, but dimensional changes in mold jigs occur leading to quality inconsistency
Solution Approach 1:
The patent incorporates sensors and detection mechanisms that monitor the positioning accuracy and dimensional state of the mold jig during operation. The robot control unit receives feedback information about the actual stacking position and compares it with the target position, then automatically adjusts the moving path and positioning parameters to compensate for any dimensional changes or wear in the mold jig, maintaining consistent quality over extended use.
Solution Approach 2:
The system dynamically adjusts operational parameters such as robot moving speed, positioning coordinates, and gripping force based on real-time conditions and accumulated usage data of the mold jig. By changing these parameters adaptively, the system compensates for dimensional changes in the mold jig over time, maintaining manufacturing consistency without requiring frequent physical recalibration of the mold jig itself.
3Ease of operation
If robot teaching methods are used for stacking, then ease of operation is improved, but stacking precision deteriorates due to predetermined deviations
Solution Approach 1:
The system combines easy-to-use robot teaching functionality with automatic feedback-based precision correction. During the teaching phase, operators can easily record basic moving paths. During actual operation, the system uses sensors to detect actual positions and automatically adjusts the paths in real-time, maintaining both ease of operation and high stacking precision by separating the simplicity of programming from the complexity of precision control.
Data Source
AI summary
A machine learning device which learns an operation of a laminated core manufacturing apparatus for stacking a plurality of core sheets to manufacture a laminated core, wherein the machine learning device includes a state observation unit which observes states of the core sheets and the laminated core manufacturing apparatus; and a learning unit which updates a manipulated variable for stacking the core sheets, on the basis of a state variable observed by the state observation unit.