Injection Molding Simulation Using Machine Learning for Real-Time Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current injection molding processes face challenges in efficiency and precision due to high computation requirements and the complexity of simulating and optimizing the injection molding process, making it time-consuming and resource-intensive, especially when trying to modify the die design to address issues during production.
Innovation Solution
A computer-implemented method that uses an external processing unit to simulate the injection molding process based on input parameters, including a simulation model and material-specific parameters, applying an optimizing algorithm to determine predicted process parameters, which are then adjusted in real-time to adapt the injection molding machine parameters until the desired properties of the workpiece are achieved within predefined tolerances, utilizing a closed-loop system that includes machine learning for continuous improvement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If injection molding simulation and optimization is performed using traditional methods, then manufacturing precision can be improved, but productivity decreases due to time-consuming simulation processes
Solution Approach 1:
The system performs preliminary simulation and optimization calculations before actual production using historical data and machine learning models to pre-determine optimal process parameters, reducing the time required during actual manufacturing operations
Solution Approach 2:
Traditional computational fluid dynamics (CFD) simulation methods are replaced with machine learning-based predictive models that can quickly estimate process parameters and outcomes without requiring intensive mechanical computation, significantly speeding up the optimization process
2Manufacturing precision
If comprehensive simulation and optimization calculations are performed, then manufacturing precision improves, but device complexity increases due to high computation requirements
Solution Approach 1:
Instead of performing full-scale complex simulations on the actual injection molding machine, the system uses simplified digital twins and virtual models that replicate the essential physics and behavior, providing accurate predictions without the computational burden of complete system simulation
Solution Approach 2:
The system transforms complex multi-parameter optimization problems into simplified parameter adjustment tasks by using machine learning models to identify the most influential parameters and focus optimization efforts only on those critical variables, reducing computational complexity while maintaining precision
3Manufacturing precision
If real-time adaptation of process parameters is implemented, then manufacturing precision improves, but device complexity increases due to additional sensors and control systems
Solution Approach 1:
The injection molding machine is equipped with self-diagnostic and self-adjustment capabilities where the system automatically monitors its own performance using integrated sensors and adjusts process parameters without external intervention, achieving real-time precision control while minimizing additional control system complexity
Solution Approach 2:
The system implements closed-loop feedback control where process parameters are continuously monitored during injection molding and automatically adjusted based on real-time measurements of workpiece properties, ensuring manufacturing precision through dynamic adaptation rather than static pre-programming
Data Source
AI summary
Disclosed herein is a computer-implemented method for controlling and/or monitoring at least one injection molding process in at least one injection molding machine. The method includes:a) providing a set of input parameters by at least one external processing unit;b) simulating an injection molding process based on the set of input parameters and determining at least one predicted process parameter of the simulated injection molding process;c) performing at least one injection molding process using the injection molding machine; andd) determining at least one actual process parameter of the injection molding process and comparing the actual process parameter and the predicted process parameter and adapting the simulation model based on the comparison.