Preform Cover Glass Shape Prediction for Curved Glass Molding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in manufacturing curved cover glass is the inability to accurately predict the shape of preform cover glass, leading to increased manufacturing costs due to extended product development periods and design of experiments.
Innovation Solution
A preform cover glass shape prediction device and method using a machine learning model that processes characteristic data from curved and flat corner parts of cover and preform glasses to generate accurate design specifications for molding target curved cover glasses, employing data preprocessing, augmentation, and synthetic data generation to improve model accuracy and reduce computational costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional manufacturing methods are used without shape prediction, then manufacturing process simplicity is maintained, but manufacturing precision deteriorates due to inability to accurately predict preform cover glass shape
Solution Approach 1:
The patent uses machine learning models to create virtual copies and simulations of the thermoforming process. Instead of physically experimenting with each preform design, the system creates digital replicas that can be tested and optimized virtually, then applies these learned patterns to predict the shape of target preform cover glasses accurately.
Solution Approach 2:
The patent performs shape prediction before actual manufacturing by training machine learning models on historical data and using them to forecast the optimal preform cover glass shape. This preliminary prediction allows manufacturers to prepare accurate designs in advance, avoiding trial-and-error during production and improving manufacturing precision.
2Manufacturing precision
If design of experiments is increased to improve shape prediction accuracy, then manufacturing precision improves, but productivity deteriorates due to extended product development periods
Solution Approach 1:
The patent replaces physical design of experiments with computational machine learning models. Instead of manually creating and testing multiple preform designs through traditional DOE methods, the system uses trained AI models to predict optimal shapes computationally, dramatically reducing development time while maintaining or improving accuracy.
Solution Approach 2:
The patent transforms the approach by changing from iterative physical experimentation to direct computational prediction. The machine learning models have learned the relationship between input parameters (curved cover glass specifications) and output parameters (preform shape characteristics), allowing direct calculation of optimal designs without time-consuming physical trials.
3Manufacturing precision
If more design of experiments is conducted to predict preform shape accurately, then manufacturing precision improves, but loss of time increases due to extended development periods
Solution Approach 1:
The patent performs comprehensive shape prediction analysis in advance by training machine learning models on extensive historical data before actual product development. This preliminary action creates a ready-to-use prediction system that can quickly determine optimal preform shapes for new designs without requiring time-consuming experimental iterations during the development phase.
Solution Approach 2:
The patent creates virtual training datasets and digital models that replicate real-world thermoforming behavior. By copying and analyzing historical manufacturing data through machine learning, the system learns optimal shape relationships without requiring additional physical experiments, thus eliminating time loss while improving prediction accuracy.
Data Source
AI summary
A preform cover glass shape prediction device for predicting a shape of a preform cover glass, the device includes a preform cover glass prediction model generator trained to obtain input data representing a training curved cover glass and a training preform cover glass, to generate cover glass characteristic data based on the input data, and to generate a predicted design specification of a target preform cover glass based on the cover glass characteristic data, and a preform cover glass shape predictor configured to generate a predicted design specification for a preform cover glass corresponding to a target curved cover glass based on the cover glass characteristic data and characteristic data of the target curved cover glass.