Steel slag asphalt interface adhesion work prediction method based on physical neural network

By constructing a prediction model for diffusion, reaction source, and multi-mechanism dissipation terms using a physical neural network-based approach, we have solved the problems of high economic cost and limited prediction accuracy in the existing technology for assessing the adhesion performance of steel slag and asphalt interfaces, and achieved higher accuracy and reliability in predicting adhesion work.

CN121999892AActive Publication Date: 2026-05-08INNER MONGOLIA UNIV OF TECH
6 Cites 2 Cited by

Patent Information

Application Number
CN202610065854.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-01-19
Publication Date
2026-05-08
Estimated Expiration
2046-01-19

AI Technical Summary

Technical Problem

Existing technologies for evaluating the interfacial adhesion performance of steel slag and asphalt suffer from high economic costs, long testing cycles, and limited prediction accuracy. In particular, they are difficult to accurately characterize the effects of complex chemical compositions and time-varying temperature under multi-field coupling.

Method used

A physical neural network-based method was adopted to construct a neural network prediction model for diffusion terms, reaction source terms, multi-mechanism dissipation terms, and state variables by acquiring the chemical composition parameters and environmental impact factors of steel slag asphalt mixture. The model was then trained in conjunction with physical constraint terms to generate prediction parameters for interfacial adhesion work.

Benefits of technology

It improves the accuracy and generalization ability of adhesion work prediction, ensures that the model output conforms to the basic physical laws of thermodynamics and mechanics, adapts to the degradation process of material properties over time, and enhances the reliability and accuracy of long-term prediction.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention proposes a steel slag asphalt interface adhesion work prediction method based on a physical neural network, and relates to the technical field of deep learning, and the method comprises the steps: obtaining chemical component parameters and environmental impact factors of a steel slag asphalt mixture interface sample; creating an original neural network prediction model based on a diffusion item, a reaction source item, a multi-mechanism dissipation item and a state variable corresponding to spatial-temporal evolution of the adhesion work; inputting the chemical component parameters and the environmental impact factors into an original neural network prediction model, and generating interface adhesion work prediction parameters in combination with physical constraint terms; training the original neural network prediction model by using the interface adhesion work prediction parameters and a gradient descent algorithm to obtain a target neural network prediction model; the physical constraint items comprise data loss, PDE residual error loss, state equation loss and boundary condition loss; and inputting the chemical component parameters and the environmental impact factors of a to-be-tested material into the target neural network prediction model to obtain target prediction parameters.
Need to check novelty before this filing date? Find Prior Art