Open Access

REVIEW

Review of the Mechanical Performance Prediction of Concrete Based on Artificial Neural Networks

Yidong Xu¹, Weijie Zhuge^1,2, Jialei Wang¹, Xiaopeng Yu^3,*, Kan Wu⁴

1 School of Civil Engineering, NingboTech University, Ningbo, 315100, China
2 School of Civil Engineering, Chongqing Jiaotong University, Chongqing, 400074, China
3 School of Mechanical and Automotive Engineering, Anhui Water Conservancy Technical College, Hefei, 231603, China
4 Zhejiang Provincial Erjian Construction Group Ltd., Ningbo, 315202, China

* Corresponding Author: Xiaopeng Yu. Email:

(This article belongs to the Special Issue: Advances in Sustainable Concrete Technologies: SCMs, Circular Economy, and AI Integration)

Structural Durability & Health Monitoring 2025, 19(6), 1507-1527. https://doi.org/10.32604/sdhm.2025.069021

Received 12 June 2025; Accepted 15 August 2025; Issue published 17 November 2025

Abstract

The performance of concrete can be affected by many factors, including the material composition, environmental conditions, and construction methods, and it is challenging to predict the performance evolution accurately. The rise of artificial intelligence provides a way to meet the above challenges. This article elaborates on research overview of artificial neural network (ANN) and its prediction for concrete strength, deformation, and durability. The focus is on the comparative analysis of the prediction accuracy for different types of neural networks. Numerous studies have shown that the prediction accuracy of ANN can meet the standards of the practical engineering applications. To further improve the applicability of ANN in concrete, the model can consider the combination of multiple algorithms and the expansion of data samples. The review can provide new research ideas for development of concrete performance prediction.

---

Keywords

---