Fluid Analysis Simulation for Injection Molding Defect Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing injection molding machine systems require exclusive use during reinforcement learning, resulting in resin material waste and prolonged learning man-hours, as they cannot efficiently adjust molding conditions to reduce defects in molded articles.
Innovation Solution
A machine learning method that simulates molding processes using a fluid analysis device to set and adjust variable parameters, acquiring defect-related parameters, and performing machine learning to output optimal variable parameters that reduce defects, thereby shortening actual molding man-hours and improving molding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If reinforcement learning is performed using actual molding machine operation, then the learning model can be trained with real molding data, but resin material is wasted and learning man-hours are prolonged
Solution Approach 1:
The patent creates a virtual copy of the molding machine through a fluid analysis device that simulates the molding process. This virtual model allows reinforcement learning to be performed using simulated resin material rather than actual resin, eliminating material waste while maintaining learning effectiveness. The simulation replicates the physical molding process conditions and outcomes without consuming physical resources.
Solution Approach 2:
The patent performs preliminary simulation actions in the virtual environment before actual molding. By pre-training the learning model using simulated data from the fluid analysis device, the system prepares optimal molding parameters in advance, reducing the need for extensive actual molding trials and thereby reducing resin waste and learning time.
2Reliability
If reinforcement learning is performed using actual molding machine operation, then the learning model can be trained with real molding data, but learning man-hours are prolonged
Solution Approach 1:
The patent creates a virtual copy of the molding machine through a fluid analysis device that simulates the molding process. This virtual model allows reinforcement learning to be performed using simulated resin material rather than actual resin, eliminating material waste while maintaining learning effectiveness. The simulation replicates the physical molding process conditions and outcomes without consuming physical resources.
Solution Approach 2:
The patent enables continuous simulation operations in the virtual environment without the interruptions inherent in physical molding. The fluid analysis device can perform repeated simulations rapidly and continuously, allowing the learning model to accumulate training data much faster than actual molding operations, thereby significantly reducing learning man-hours.
3Productivity
If simulation is used to reduce actual molding operations, then resin waste and learning time are reduced, but the accuracy of learning data may be compromised
Solution Approach 1:
The patent implements a feedback mechanism where the learning model is trained to predict both molding parameters and defect outcomes based on simulation data. The fluid analysis device provides detailed feedback on virtual molding results, including defect generation conditions, which are used to continuously refine the learning model's accuracy. This feedback loop ensures that simulated training data effectively translates to accurate real-world predictions.
Solution Approach 2:
The patent utilizes the fluid analysis device's ability to vary multiple molding parameters systematically in simulation. By changing parameters such as injection pressure, temperature, and speed in the virtual environment, the learning model learns the relationships between parameter adjustments and defect outcomes. This comprehensive parameter exploration in simulation builds accurate predictive capabilities without requiring extensive physical trials.
Data Source
AI summary
Provided is a machine learning method of a learning model that outputs a variable parameter that is configured to reduce the degree of defect of a molded article obtained by actual molding and relates to molding conditions of a molding machine in a case where observation data obtained by observing a physical quantity relating to actual molding using the molding machine is input. The machine learning method includes: a step of simulating a molding process by setting a variable parameter and a fixed parameter to a fluid analysis device; a step of acquiring a defect-related parameter that is obtained by simulation and relates to the degree of defect of the molded article; a step of calculating the degree of defect of the molded article on the basis of the acquired defect-related parameter; and a step of causing the learning model to perform machine learning by using the variable parameter set to the fluid analysis device and reward corresponding to the calculated degree of defect.