Composite Material Estimation Using Numerical Characteristic Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for estimating the physical quantity of composite materials face challenges in accuracy when new constituent materials not used in the training data are included, due to the use of material names as input parameters which are difficult to synthesize, leading to reduced estimation accuracy.
Innovation Solution
The system generates an approximate function based on characteristic values and blending ratios of constituent materials, allowing for the estimation of physical quantities even when new materials are involved, by using numerical values as input parameters, thereby improving estimation accuracy and expanding the range of applicable composite materials.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If material names are used as input parameters for estimating physical quantity, then the system can handle known materials in training data, but estimation accuracy deteriorates when new constituent materials not used in training data are included
Solution Approach 1:
The patent transforms the input parameters from material names (categorical data) to characteristic values (numerical data). This parameter transformation allows the system to handle both known and new materials effectively, as numerical values can be synthesized through mathematical operations even for materials not present in the training data, thereby maintaining estimation accuracy while expanding adaptability.
Solution Approach 2:
The patent introduces characteristic values as an intermediary between material names and the estimation model. Instead of directly using material names as inputs, the system converts them to numerical characteristic values that capture essential material properties. This intermediary representation enables the model to process both familiar and novel materials through numerical synthesis operations.
2Ease of manufacture
If material names are used as input parameters, then the system is simple to implement, but the input parameters become difficult to synthesize for new materials
Solution Approach 1:
By changing the parameter type from material names to characteristic values, the patent simplifies the synthesis process. Numerical characteristic values can be combined using mathematical operations (weighted averages based on blending ratios) to create synthesized characteristic values for composite materials, whereas material names cannot be synthesized through such operations.
3Measurement precision
If characteristic values and blending ratios are used to generate approximate functions, then estimation accuracy improves for new materials, but the calculation process becomes more complex
Solution Approach 1:
The patent performs preliminary conversion of material names to characteristic values before the estimation process. By pre-processing the input data into numerical characteristic values and pre-calculating synthesized characteristic values based on blending ratios, the system reduces the complexity during the actual estimation phase, making the overall process more manageable despite the additional initial steps.
Data Source
AI summary
A first synthesized characteristic value calculated based on a first blending ratio of constituent materials contained in a first composite material and on a first characteristic value corresponding to the constituent materials contained in the first composite material, as well as first blending information of the constituent materials contained in the first composite material are used as input parameters (explanatory variables) for an approximate function to estimate a value of physical quantity (objective variable) for the first composite material.