Thermal Analysis Signal Evaluation Using Sliding-Window AI Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing thermal analysis methods, such as differential scanning calorimetry (DSC), rely heavily on manual evaluation of measurement signals, which is time-consuming and lacks automation, and existing automated methods often require complex neural networks, making them less accessible and efficient.

Innovation Solution

A method using an artificial intelligence module with support vector machines and random forest methods to classify thermal effects in thermal analysis data by applying sliding windows to measurement curves, allowing for automated and precise detection of thermal effects like glass transitions and melting points without the need for neural networks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If manual evaluation methods are used for thermal analysis measurement signals, then evaluation accuracy can be maintained through expert judgment, but the process is time-consuming and lacks automation

Engineering Contradiction:
Improveautomation of thermal analysis evaluationVSAvoidevaluation time
Core Design Contradiction:
Extent of automationVSLoss of time

Solution Approach 1:

The system enables automated self-evaluation of thermal analysis measurement signals through classification algorithms that automatically identify and characterize thermal effects without requiring manual expert intervention, thereby achieving both automation and time efficiency

Inventive Principle:
Principle #25Self-service

2Measurement precision

If complex neural network architectures are used for automated thermal effect detection, then detection accuracy can be improved, but the device complexity and computational requirements increase

Engineering Contradiction:
Improvethermal effect detection accuracyVSAvoidcomplexity of AI module
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent changes the parameter of the AI approach from complex neural networks to simpler classification algorithms such as support vector machines and random forest methods, achieving good detection accuracy with reduced computational complexity and easier implementation

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent employs computationally lighter classification algorithms that are easier to implement and require fewer computational resources compared to complex neural networks, making the system more accessible and efficient for routine thermal analysis evaluation

Inventive Principle:
Principle #27Cheap short-living objects (Disposable)

3Measurement precision

If the entire measurement curve is analyzed at once, then comprehensive thermal effect detection is achieved, but the classification process becomes computationally intensive and less precise

Engineering Contradiction:
Improvethermal effect boundary determination precisionVSAvoidcomplexity of classification process
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the measurement curve into multiple overlapping sliding windows, allowing the classification algorithm to process smaller segments individually. This segmentation improves precision by focusing on local thermal effect boundaries while reducing the computational complexity of analyzing the entire curve at once

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4711751A1Method and device for evaluating a measurement signal of thermal analysis
Publication Date: 2026.03.18 NETZSCH GERATEBAU GMBH
  • EP4711751A1 patent drawingFigure 1~2
  • EP4711751A1 patent drawingFigure 3~4
  • EP4711751A1 patent drawingFigure 5

AI summary

The invention relates to a method and a device (100) for evaluating a measurement signal from a thermal analysis. The device (100) has a data interface (110) configured to receive the measurement signal from the thermal analysis, wherein the measurement signal represents a measurement curve based on a temperature series, and a processing logic (120).The processing logic (120) is configured to determine a number of sliding windows based on the measurement signal, wherein each sliding window is assigned to a corresponding section of the measurement curve with a number of measurement points, and to determine, by means of an artificial intelligence module executed by the processing logic and configured for classification, whether a thermal effect of one of the sample material underlying the thermal analysis is present for the respective number of sliding windows, wherein the artificial intelligence module is configured to determine a continuous section of the thermal effect with respect to the measurement curve.