Surface Topography Reconstruction for Accurate Property Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for surface topography reconstruction, such as those based on statistical methods and spatial frequency analysis, fail to accurately predict the performance of surfaces due to a lack of physical explanation and the difficulty in extracting characteristic parameters, leading to low accuracy in predicting properties like current-carrying performance and curvature sensitivity.

Innovation Solution

A method integrating machine learning and spatial frequency analysis to generate two-dimensional arrays, extract characteristic variables, and establish a relationship between these variables and surface properties for performance prediction, using a formula to generate surface topography and a regression model to improve accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If statistical methods are used for surface topography reconstruction, then the reconstruction process is simple, but the accuracy in predicting current-carrying performance and curvature-sensitive properties is low

Engineering Contradiction:

Improvereconstruction process simplicityVSAvoidprediction accuracy of current-carrying performance

Core Design Contradiction:

Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent combines statistical methods with spatial frequency analysis and machine learning algorithms to create a composite reconstruction approach. This integrates the simplicity of statistical methods with the predictive power of spatial frequency analysis and ML models, achieving both ease of manufacture and high prediction accuracy for current-carrying performance and curvature-sensitive properties.

Inventive Principle:

Principle #40Composite materials

Solution Approach 2:

The patent introduces spatial frequency analysis as an intermediary between statistical methods and machine learning. This intermediary layer extracts meaningful features from surface profiles, bridging the gap between simple statistical reconstruction and complex ML-based prediction, thereby improving accuracy without completely abandoning the simplicity of statistical approaches.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If spatial frequency analysis method is used for surface reconstruction, then the reconstruction accuracy can be improved, but the difficulty in extracting characteristic parameters from actual rough surfaces increases

Engineering Contradiction:

Improvereconstruction accuracyVSAvoidparameter extraction difficulty

Core Design Contradiction:

Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent employs machine learning algorithms that automatically extract characteristic parameters from actual rough surface data. The ML models perform self-service by learning the complex relationships between spatial frequency components and surface properties, eliminating the need for manual parameter extraction and reducing the difficulty of detecting and measuring characteristic parameters.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical analysis of spatial frequency components with automated machine learning systems. This substitution transforms the complex manual extraction process into an automated computational approach, significantly reducing the difficulty of detecting and measuring characteristic parameters while maintaining high reconstruction accuracy.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If machine learning and spatial frequency analysis are integrated, then the prediction accuracy of surface properties is improved, but the device complexity increases

Engineering Contradiction:

Improveprediction accuracy of surface propertiesVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the surface analysis process into distinct modules: spatial frequency analysis module, machine learning prediction module, and parameter extraction module. This segmentation allows each component to be optimized independently and simplifies the overall system by dividing complexity into manageable, functional units that work together to achieve high prediction accuracy.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS20260004012A1Surface topography reconstruction method and system integrating machine learning and spatial frequency analysis

Publication Date: 2026.01.01 SHANGHAI JIAOTONG UNIV

AI summary

A surface topography reconstruction method integrating machine learning and spatial frequency analysis includes step 1: Using the analytical formula of the spatial frequency method to generate a large number of two-dimensional arrays that symbolize rough surfaces; step 2: by analyzing the height distribution and the frequency energy ratio of the generated surface, combined with the machine learning method, extracting the characteristic variable of the analytical formula; step 3: integrating the test or the simulation method to establish the relationship between the characteristic parameter and the surface property for performance prediction and surface reconstruction in a specific scenario. The surface topography reconstruction method can realize the surface topography characteristic extraction based on the spatial frequency analysis method, and reconstruct the rough surface suitable for the current-carrying performance prediction, thereby improving the practical significance of the surface reconstruction method in performance prediction.