Method and apparatus for predicting the physical properties of resin compositions
A machine learning-based method and apparatus predict resin composition properties by correlating blending amounts and material characteristics, addressing inaccuracies in existing methods and enhancing prediction accuracy for elongation and tensile strength.
JP7882059B2Active Publication Date: 2026-06-30PROTERIAL LTD
Patent Information
- Authority / Receiving Office
- JP Β· JP
- Patent Type
- Patents
- Current Assignee / Owner
- PROTERIAL LTD
- Filing Date
- 2022-09-02
- Publication Date
- 2026-06-30
AI Technical Summary
Technical Problem
Existing methods struggle to accurately predict the elongation and tensile strength of non-halogen-based resin compositions used as coating materials for electric wires, despite knowing the blending amounts of materials.
Method used
A method and apparatus that utilize machine learning to create a regression model based on filler surface area, maleic anhydride modification ratio, filler volume ratio, crystallographic data, and vinyl acetate group amount data to predict elongation and tensile strength by correlating blending amounts and material characteristics.
Benefits of technology
Accurately predicts the elongation and tensile strength of resin compositions, improving prediction accuracy by selecting appropriate material characteristic data for specific properties.
β¦ Generated by Eureka AI based on patent content.
Smart Images
Figure 0007882059000001 
Figure 0007882059000002 
Figure 0007882059000003
Abstract
To provide a method and a device for predicting physical property of a resin composition which can accurately predict elongation and tensile strength of a resin composition.SOLUTION: A physical property prediction method of a resin composition, in which physical property as a prediction object is elongation and tensile strength of a resin composition, machine-learns a relation among blending amount data 11 including information on a blending amount of a material, material feature data 12 including at least one data of (a) filler surface area data, (b) anhydride maleic acid modified ratio data, (c) filler volume fraction data, (d) crystal amount data and (e) vinyl acetate group amount data, and physical property data 13 indicating physical property of the resin composition, creates a regression model 7 indicating a correlation among the blending amount data 11, the material feature data 12, and the physical property data 13, and predicts the physical property data 13 according to the blending amount data 11 and the material feature data 12 as prediction sources, using the regression model 7.SELECTED DRAWING: Figure 3
Need to check novelty before this filing date? Find Prior Art