Scholarlink

<p>The two most important mechanical properties for concrete are compressive strength and static modulus. Likewise, Wenner resistivity is a crucial durability parameter to&#xa0;be taken into consideration while monitoring the performance of any concrete members. This paper presents novel prediction models for normal concrete’s compressive strength, static modulus, and Wenner resistivity based on linear regression models and artificial neural networks (ANN). Due to the quicker rate of output convergence, the study used the Levenberg–Marquardt learning algorithm for the ANN model to forecast the aforementioned parameters. The prediction strength (R2) of the ANN technique is 14–20% higher than that of the normal regression model, 11–14% higher than that of the static modulus model, and 10–12.5% higher than that of the Wenner resistivity model. For both ANN and linear regression models, the input parameters considered were the rebound number and pulse velocity. The sample was evaluated by substituting normal stone aggregate (NSA) with varying amounts of recycled concrete aggregate (i.e., 0%, 25%, 50%, 75%, and 100% RCA) as a coarse aggregate. This study considered age (14, 28, and 90&#xa0;days) and grade (M20, M25, and M30) into consideration while developing the models. Furthermore, by comparing the developed compressive strength model with earlier models created by other authors, the study found that the generated model performed better for RCA specimens. The findings of this investigation will support the application of RCA in the Indian construction sector and promote utilization of natural coarse aggregate more sustainably.</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

Prediction of compressive strength, static modulus and wenner resistivity for normal concrete using different percentages of recycled concrete as a coarse aggregate

  • 期刊论文

  • 发表日期：

    2025年4月1日

查看全文
  • Sheetal Thapa,
  • Nagondanahalli Raju Asha Rani,
  • Richi Prasad Sharma

摘要

The two most important mechanical properties for concrete are compressive strength and static modulus. Likewise, Wenner resistivity is a crucial durability parameter to be taken into consideration while monitoring the performance of any concrete members. This paper presents novel prediction models for normal concrete’s compressive strength, static modulus, and Wenner resistivity based on linear regression models and artificial neural networks (ANN). Due to the quicker rate of output convergence, the study used the Levenberg–Marquardt learning algorithm for the ANN model to forecast the aforementioned parameters. The prediction strength (R2) of the ANN technique is 14–20% higher than that of the normal regression model, 11–14% higher than that of the static modulus model, and 10–12.5% higher than that of the Wenner resistivity model. For both ANN and linear regression models, the input parameters considered were the rebound number and pulse velocity. The sample was evaluated by substituting normal stone aggregate (NSA) with varying amounts of recycled concrete aggregate (i.e., 0%, 25%, 50%, 75%, and 100% RCA) as a coarse aggregate. This study considered age (14, 28, and 90 days) and grade (M20, M25, and M30) into consideration while developing the models. Furthermore, by comparing the developed compressive strength model with earlier models created by other authors, the study found that the generated model performed better for RCA specimens. The findings of this investigation will support the application of RCA in the Indian construction sector and promote utilization of natural coarse aggregate more sustainably.