Polymer Design Device Using Regression Model for Property Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Traditional polymer design methods are inefficient and unreliable, relying on random selection and experimentation to achieve desired physical properties, making the development of new functional materials time-consuming and costly.

Innovation Solution

A polymer design device utilizing deep reinforcement learning to predict polymer structures with specific physical properties by creating a regression model based on structural information and physical property data, incorporating a score evaluation system to select optimal polymers based on target properties and their standard deviations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If random selection and experimentation methods are used for polymer design, then the process is simple to implement, but the efficiency and reliability of finding polymers with desired physical properties deteriorates

Engineering Contradiction:

Improveease of implementationVSAvoidefficiency of polymer design

Core Design Contradiction:

Ease of manufactureVSProductivity

Solution Approach 1:

The patent replaces the mechanical/random experimentation system with an information processing system using machine learning. The polymer design device uses a regression model to predict physical properties from structural information, substituting random trial-and-error with data-driven prediction, thereby dramatically improving efficiency while maintaining ease of use through automated processing

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces a regression model as an intermediary between polymer structural information and physical properties. This model acts as a mediator that predicts physical properties without requiring actual experimentation, enabling efficient screening of polymers with desired properties while reducing the need for time-consuming physical tests

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of manufacture

If random selection and experimentation methods are used for polymer design, then the implementation process is simple, but the reliability of predicting polymer structure deteriorates

Engineering Contradiction:

Improveease of implementationVSAvoidreliability of polymer structure prediction

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The patent replaces unreliable random experimentation with a machine learning-based regression model that learns from training data. This substitution transforms the prediction process from random guessing to systematic data-driven analysis, significantly improving reliability while keeping the system easy to use through automated model application

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent implements a feedback mechanism where the regression model is trained on known polymer structure-property relationships and uses this learned knowledge to make predictions. The model continuously improves its reliability by incorporating training data feedback, enabling accurate prediction of polymer structures with desired physical properties

Inventive Principle:

Principle #23Feedback

3Device complexity

If traditional experimentation methods are used to develop new functional materials, then no additional equipment is needed, but the time and cost required deteriorates

Engineering Contradiction:

Improveequipment requirementsVSAvoiddevelopment time

Core Design Contradiction:

Device complexityVSLoss of time

Solution Approach 1:

The patent substitutes physical experimentation with computational prediction using a regression model. This replacement eliminates the need for time-consuming synthesis and testing cycles, dramatically reducing development time while requiring only standard computing equipment and existing polymer data, thus maintaining acceptable device complexity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent performs preliminary computational screening and prediction before actual polymer synthesis. By using the regression model to pre-evaluate potential polymers and identify promising candidates, the system reduces the number of actual experiments needed, thereby reducing development time without requiring complex additional equipment

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11126695B2Polymer design device, polymer design method, and non-transitory recording medium

Publication Date: 2021.09.21 RESONAC CORP

AI summary

A polymer design device according to an embodiment of the present disclosure receives a requirement for a target physical property of a desired polymer, and acquires structural information of polymers. For each polymer corresponding to the acquired structural information, the polymer design device estimates physical property information of the polymer including a mean value and a standard deviation, based on the structural information of the polymer and a regression model, and calculates a score of the polymer based on the requirement for the target physical property and based on the mean value and the standard deviation. From among the acquired structural information of the polymers, the polymer design device selects at least one polymer as the desired polymer, based on the score of each of the polymers, and outputs information of the selected at least one polymer.