Material Phase-Ratio Prediction Heat Maps for Alloy Design

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The high workload and difficulty for designers to understand the impact of multiple additive elements on phase ratios in alloys make it challenging to efficiently design materials with desired phase ratios.

Innovation Solution

A prediction device using a trained model to predict phase ratios at each temperature within a predetermined range, applying architectures like RNN, bidirectional RNN, Seq2Seq, Seq2Seq with Attention mechanism, GRU, LSTM, or Transformer, and displaying the results in a heat map to facilitate understanding of the impact of material composition modifications.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If the phase ratio is predicted by successively increasing or decreasing the weight % of all additive elements and comparing with the phase ratio before modification, then the degree of impact of each additive element on the phase ratio can be understood, but the work load on the designer increases and it becomes difficult to understand the degree of impact when the alloy includes a large number of additive elements

Engineering Contradiction:
ImproveUnderstanding of impact degreeVSAvoidWork load
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent extracts only the necessary information (degree of impact of each additive element) from the complex prediction process and presents it in a simplified format (heat map) that eliminates the need for designers to manually perform successive predictions and comparisons of all additive elements

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent uses color-coded heat maps to visually represent the degree of impact, where different colors indicate different levels of impact. This allows designers to quickly comprehend the effect of each additive element without having to analyze numerical data from multiple successive predictions

Inventive Principle:
Principle #32Color changes

2Loss of information

If the phase ratio prediction is performed for a large number of additive elements by successively modifying each element's weight %, then complete information about impact degree is obtained, but the time required for prediction and analysis increases significantly

Engineering Contradiction:
ImproveComplete impact informationVSAvoidPrediction and analysis time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The system performs preliminary calculations of the degree of impact for all additive elements in the background process before presenting the results to the designer. This allows complete impact information to be prepared in advance, eliminating the need for designers to perform time-consuming successive predictions during the design process

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a visual copy (heat map) of the complex prediction results that can be quickly displayed and analyzed. Instead of requiring designers to examine multiple prediction outputs and comparisons, the system generates a single visual representation that captures all necessary impact information

Inventive Principle:
Principle #26Copying

Data Source

PatentEP4645324A1Prediction device, prediction method, and prediction program
Publication Date: 2025.11.05 RESONAC CORP
  • EP4645324A1 patent drawingFigure 1
  • EP4645324A1 patent drawingFigure 2
  • EP4645324A1 patent drawingFigure 3

AI summary

One object is to easily understand the degree of impact on a phase ratio caused by a modification of a material composition. A prediction device includes a storage unit configured to store a trained model that is trained using training data in which a material composition of a material to be trained, and a phase ratio of the material to be trained at each temperature within a predetermined temperature range, are associated with each other, wherein a value predicted for a phase specified in advance is partially differentiated using a material composition of a material to be predicted at each temperature, when the phase ratio of the material to be predicted at each temperature within the predetermined temperature range is predicted by inputting the material composition of the material to be predicted to the trained model, and each area specified based on each component and based on each temperature is displayed in a display mode according to the partially differentiated value, in a heat map in which each component of the material to be predicted is arranged on a first axis and each temperature within the predetermined temperature range is arranged on a second axis.