AI Composition-Process Model for Composite Materials
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Developing a method to predict the optimal composition and production process for composite materials that satisfy specific properties is challenging due to the high cost and time required for experimental trials, leading to low accuracy and reliability in results.
Innovation Solution
An AI-based device and method that collects and classifies composition-process condition data, generates a composition-process model, and derives optimal conditions for a target property, utilizing machine learning with online data sources like papers and patents to recommend composition and process variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If experimental trials are conducted to find optimal composition and process for composite materials, then accuracy and reliability of results improve, but cost and time requirements increase significantly
Solution Approach 1:
The system performs preliminary data collection and model training before actual composition optimization is needed. Historical composition data and process parameters are gathered in advance, and AI models are pre-trained to predict optimal compositions, eliminating the need for extensive experimental trials when optimization is required.
Solution Approach 2:
Instead of conducting physical experimental trials, the system creates virtual copies of experimental conditions through AI simulation. The trained models generate predicted composition recommendations that replicate the outcomes of physical experiments without requiring actual material synthesis and testing.
2Reliability
If experimental trials are conducted to find optimal composition and process for composite materials, then accuracy and reliability of results improve, but cost increases significantly
Solution Approach 1:
The system replaces expensive physical experiments with AI-based virtual simulations. By copying experimental conditions and outcomes through trained models, the system achieves reliable composition recommendations without the material costs, equipment expenses, and resource consumption associated with physical trials.
Solution Approach 2:
The patent substitutes the mechanical and chemical experimental process with an information-processing system. Instead of physically mixing materials and testing properties, the system uses AI algorithms to analyze composition data and predict optimal formulations, replacing the entire experimental apparatus with computational methods.
3Quantity of substance
If numerous trials and errors are performed to find optimal composition, then comprehensive data is gathered, but accuracy and reliability of results decrease
Solution Approach 1:
The system incorporates feedback mechanisms where AI models learn from historical composition data and experimental results, continuously improving prediction accuracy. The model is trained on comprehensive datasets and refined through iterative learning, ensuring that recommendations become increasingly reliable with each training cycle.
Solution Approach 2:
The system transforms the approach from varying compositions through numerous trials to systematically analyzing composition parameters using AI. The model evaluates multiple composition parameters simultaneously and identifies optimal combinations through computational analysis rather than random experimentation, improving both efficiency and reliability.
Data Source
AI summary
The present invention relates to an artificial intelligence-based device comprising a data collection unit configured to collect composition-process condition data for a target property input by a user and store the collected condition data in a collection database; an input grade classification unit configured to classify the collected condition data into different input grades according to an input grade determination factor; a training data supply unit configured to store the condition data classified into the input grades in a training database and input condition data of a predetermined high grade in the training database; a model generation unit configured to learn and verify the data input from the training data supply unit and generate a composition-process model; and a data output unit configured to derive one or more composition-process conditions for the target property and store the derived composition-process conditions in an output database.