Cement-emulsified asphalt mixture shrinkage behavior prediction method based on deep learning

Abstract

The invention belongs to the technical field of civil engineering, and discloses a cement-emulsified asphalt mixture shrinkage behavior prediction method based on deep learning. The method comprises the steps: determining an image generation design parameter and a shrinkage performance design parameter; taking the designed cement-emulsified asphalt mixture data packet as to-be-trained data, and constructing an image generation training sample set; constructing and training a generative adversarial neural network model; constructing and training a shrinkage performance deep neural network; andgiving design parameters of a to-be-predicted mixture, and predicting the shrinkage rate of the target mixture by adopting the trained generative adversarial network and the shrinkage performance prediction model. According to the method, the shrinkage performance of the finally designed product can be accurately predicted through the design parameters, the design proportion of the cement-emulsified asphalt mixture can be rapidly optimized, and the method has positive significance in improving the shrinkage performance of the mixture.

Description

technical field [0001] The invention belongs to the technical field of civil engineering materials, and in particular relates to a method for predicting shrinkage behavior of cement emulsified asphalt mixture based on deep learning. Background technique [0002] Cement emulsified asphalt mixture is a heterogeneous system composed of cement and emulsified asphalt, coarse and fine aggregates, and internal voids; the water in the mixture is in a free state and occupies a large number of voids in the dispersion system of the mixture. Finally, voids will be generated inside the mixture, resulting in poor volume stability and obvious environmental impact on durability, which is manifested in the poor overall uniformity of the mixture, large void ratio and serious shrinkage cracking. [0003] Accurate detection or prediction of shrinkage of cement emulsified asphalt mixture is very important for the use of this type of material. At present, the detection methods for the shrinkage ...

AI Technical Summary

Problems solved by technology

However, these test methods are cumbersome to operate, require high technical requirements, and the test period is long, especially the detection accuracy is low; and the low-precision test results will lead to wrong guidance of material design parameters, which will seriously affect the performance of the design material

[0004] Traditional microscopic detection methods only carry out weak correlation and one-sided subjective qualitative analysis of the local microscopic characteristics and macroscopic properties of the mixture manually.

Images