Polymer Composite Property Prediction from Material Recipes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting the properties of polymer composite materials are inefficient and costly, relying on trial and error and requiring extensive laboratory testing, and existing AI models struggle to account for the numerous variables and interactions in material recipes.
Innovation Solution
A method and device using a property prediction model trained on a dataset of learning recipes, allowing for the rapid prediction of polymer composite material properties based on input materials and mixing ratios, utilizing deep learning algorithms to infer properties and attributes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If trial and error methods with extensive laboratory testing are used to predict polymer composite material properties, then measurement precision can be improved, but loss of time and loss of substance increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-training an AI model with extensive material property data and interaction mechanisms before actual prediction tasks. This pre-processing of knowledge allows the system to quickly predict polymer composite properties without performing repeated laboratory experiments, thereby reducing development time while maintaining prediction accuracy.
Solution Approach 2:
The patent uses copying by creating a virtual model of the polymer composite material system that replicates real-world material behaviors. The AI model copies and simulates the complex interactions between materials and processing conditions, allowing predictions to be made through simulation rather than physical experimentation, thus saving time and resources.
2Measurement precision
If trial and error methods with extensive laboratory testing are used to predict polymer composite material properties, then measurement precision can be improved, but loss of substance increases significantly
Solution Approach 1:
The patent uses copying by creating a virtual model of the polymer composite material system that replicates real-world material behaviors. The AI model copies and simulates the complex interactions between materials and processing conditions, allowing predictions to be made through simulation rather than physical experimentation, thus saving time and resources.
Solution Approach 2:
The patent replaces the mechanical system of physical laboratory testing with an information-processing system based on AI and big data. Instead of physically mixing and testing materials, the system uses computational models to predict properties, substituting material consumption with data processing.
3Productivity
If commercially available artificial intelligence models are used for prediction, then productivity can be improved, but device complexity increases due to training requirements
Solution Approach 1:
The patent applies preliminary action by pre-training the AI model with extensive material property data and interaction mechanisms before actual prediction tasks. This pre-processing of knowledge allows the system to quickly predict polymer composite properties without performing repeated laboratory experiments, thereby reducing development time while maintaining prediction accuracy.
4Productivity
If existing AI models are used for prediction, then productivity can be improved, but measurement precision deteriorates due to inability to consider numerous variables and interactions
Solution Approach 1:
The patent applies parameter changes by incorporating numerous material-specific parameters and their interactions into the AI model. The system adjusts and optimizes multiple parameters simultaneously, including material compositions, ratios, and processing conditions, to accurately predict polymer composite properties while maintaining high prediction speed.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
According to various embodiments of the present disclosure, a method for predicting characteristics of a polymer composite material and a device thereof may be provided, wherein the method comprises: inputting a recipe including two or more materials containing at least one polymer and a mixing ratio for each of the two or more materials; predicting properties of the polymer composite material according to the recipe based on a recipe and property prediction model; and outputting the properties of the polymer composite material.