Scholarlink

<p>Concrete’s strength and durability (S&amp;D) make it a widely used construction material; however, water loss due to evaporation can limit proper hydration, reducing performance. This study investigates a self-curing (SC) approach utilizing polyethylene glycol-400 (PEG-400) and a Similarity-Navigated Graph Neural Network (SNGNN) for optimal dosage prediction. PEG-400 acts as an internal curing agent (ICA) by retaining moisture and improving hydration. The SNGNN method predicts the ideal PEG-400 content based on desired properties, including compressive strength, setting time, and durability. Experiments on M20, M25, and M30 concretes with 1–2% PEG-400 showed an increase in 7-day compressive strength (CS) from 33 MPa to 36 MPa, and a corresponding increase in 28-day strength from 53 MPa to 61 MPa. Split tensile and flexural strengths also improved, along with workability and compaction. Durability tests demonstrated reduced weight loss (5.3% to 2.63%) and lower compressive strength reduction (12.24% to 7.5%). Overall, 1–1.5% PEG-400 was identified as optimal, enhancing concrete strength, durability, and self-curing efficiency without external water curing. The results demonstrate improved short-term durability and mechanical performance with PEG-400. The proposed SNGNN model outperforms K-Nearest Neighbors (KNN), Feed‑forward Back Propagation Network (FFBPN), and Artificial Neural Network (ANN), achieving 93.8% accuracy, a 1.8 MPa mean absolute error, and a computation efficiency of 1.2 s, demonstrating high precision and efficiency.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Innovative Self-Curing Concrete Leveraging the Hydration-Enhancing Properties of Polyethylene Glycol-400 for Improved Strength and Durability in Construction Applications

  • 期刊论文

  • 发表日期：

    2025年11月4日

查看全文
  • Mahesh Kona,
  • Yashwanth Pamu,
  • Venkata Sarath Pamu,
  • Praveen Samarthi

摘要

Concrete’s strength and durability (S&D) make it a widely used construction material; however, water loss due to evaporation can limit proper hydration, reducing performance. This study investigates a self-curing (SC) approach utilizing polyethylene glycol-400 (PEG-400) and a Similarity-Navigated Graph Neural Network (SNGNN) for optimal dosage prediction. PEG-400 acts as an internal curing agent (ICA) by retaining moisture and improving hydration. The SNGNN method predicts the ideal PEG-400 content based on desired properties, including compressive strength, setting time, and durability. Experiments on M20, M25, and M30 concretes with 1–2% PEG-400 showed an increase in 7-day compressive strength (CS) from 33 MPa to 36 MPa, and a corresponding increase in 28-day strength from 53 MPa to 61 MPa. Split tensile and flexural strengths also improved, along with workability and compaction. Durability tests demonstrated reduced weight loss (5.3% to 2.63%) and lower compressive strength reduction (12.24% to 7.5%). Overall, 1–1.5% PEG-400 was identified as optimal, enhancing concrete strength, durability, and self-curing efficiency without external water curing. The results demonstrate improved short-term durability and mechanical performance with PEG-400. The proposed SNGNN model outperforms K-Nearest Neighbors (KNN), Feed‑forward Back Propagation Network (FFBPN), and Artificial Neural Network (ANN), achieving 93.8% accuracy, a 1.8 MPa mean absolute error, and a computation efficiency of 1.2 s, demonstrating high precision and efficiency.