Machine Learning Model Generation for Resin Composition Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional methods for finding the required characteristic satisfying condition for resin compositions containing inorganic filling materials and resins are inefficient, often requiring numerous trials and errors to achieve desired properties.

Innovation Solution

A model generation device and prediction device that utilize machine learning to generate and predict the required characteristic satisfying conditions by acquiring and processing input data related to inorganic filling material and resin characteristics, as well as their mixing ratios, to recommend optimal compositions that satisfy specific resin composition characteristics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional trial-and-error experimental methods are used to find resin composition conditions, then the required characteristic satisfying condition can be found, but the process requires numerous trials and errors which is inefficient and time-consuming

Engineering Contradiction:

Improveaccuracy of finding required characteristic satisfying conditionVSAvoidtime required for trials and errors

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary actions by acquiring resin composition data, inorganic filling material data, and resin data in advance, and generating a prediction model before actual composition design. This preliminary model generation enables efficient prediction without requiring trial-and-error experiments later

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces the mechanical trial-and-error experimental system with an information processing system. Instead of physically mixing and testing numerous compositions, the system uses data acquisition, machine learning model generation, and prediction algorithms to determine optimal compositions computationally

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Manufacturing precision

If numerous experimental trials are conducted to determine resin composition, then the desired properties can be achieved, but production costs increase due to extensive experimentation

Engineering Contradiction:

Improveachievement of desired resin propertiesVSAvoidproduction costs

Core Design Contradiction:

Manufacturing precisionVSLoss of energy

Solution Approach 1:

The system creates a virtual copy of the resin composition development process through data acquisition and prediction modeling. Instead of physically conducting numerous expensive experiments, the system uses computational models that replicate the relationship between composition and properties, allowing cost-effective determination of optimal formulations

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system efficiently explores parameter space by using the prediction model to evaluate different composition parameters (inorganic filling material types, resin types, mixing ratios) without physical experimentation. This allows systematic optimization of parameters at minimal cost

Inventive Principle:

Principle #35Parameter changes

3Loss of information

If traditional experimental methods are used to optimize resin composition, then sufficient data can be obtained, but the process is complex and requires multiple steps of data collection and analysis

Engineering Contradiction:

Improvecompleteness of composition dataVSAvoidcomplexity of data collection and processing system

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The system merges multiple data acquisition functions into a unified information processing framework. The first acquisition unit collects resin composition data, the second acquires inorganic filling material data, and the third acquires resin data, all integrated into a single prediction model generation process, simplifying the overall complexity

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS20240311692A1Model generation device, prediction device, model generation method, prediction method, and resin composition manufacturing system

Publication Date: 2024.09.19 TOKUYAMA CORP

AI summary

In order to find a required characteristic satisfying condition regarding a resin composition more efficiently than a conventional technology, a model generation device (100) is configured such that: a first machine learning section (21) generates, on the basis of first input data (110) and second input data (120) that makes a pair with the first input data (110), a first prediction model (MODEL 1) that predicts unknown resin composition characteristic data from at least one selected from the group consisting of (i) given inorganic filling material characteristic data, (ii) given resin characteristic data, (iii) given inorganic filling material proportion data and (iv) given resin proportion data; and a second machine learning section (22) generates, on the basis of the first prediction model (MODEL 1), a second prediction model (MODEL 2) that predicts at least one selected from the group consisting of (i) predicted inorganic filling material characteristic data, (ii) predicted resin characteristic data, (iii) predicted inorganic filling material proportion data and (iv) predicted resin proportion data, the at least one satisfying given resin composition characteristic data.