Dr. R. Premsudha, Dr. S. Kapilan, G. Viswanathan, M. Naganathan, S. Dhamodaran
{"title":"Concrete Breaking Strength Prediction Using Machine Learning","authors":"Dr. R. Premsudha, Dr. S. Kapilan, G. Viswanathan, M. Naganathan, S. Dhamodaran","doi":"10.48175/ijarsct-18237","DOIUrl":null,"url":null,"abstract":"When it comes to estimating, classifying, and forecasting material strength based on changing material parameters, machine learning (ML) techniques have shown to be dependable methodologies. It is found that choosing the right machine learning technique depends on the characteristics of the problem and the available data. Therefore, fifteen different machine learning techniques were used to a specific dataset of concrete compressive strength in order to assess the accuracy of ML models to predict concrete compressive strength. Due to its excellent performance while dealing with continuous target variables and nonlinear interactions among the features and the target, the Support Vector Regressor (SVR) had the greatest prediction accuracy (88.18%) of all the ML methods employed. To guarantee the structural integrity of building projects, it is essential to predict the breaking strength of concrete. The goal of this project is to create a machine learning model that can forecast concrete's breaking strength depending on the mix's composition and curing circumstances. A dataset was created that included details regarding concrete samples, such as mix ratios, curing temperatures, curing times, and breaking strengths. recise estimation of concrete's compressive strength is crucial for the advancement and construction. A bibliometric analysis of the pertinent literature published in was conducted in order to comprehend the state of research in the field of concrete compressive strength prediction. The previous ten years have seen the first research in this sector. The database consisted of 31,35 journal articles published between 2012 and 2021 in the Web of Science core database. The knowledge map was created using Cite Space 6.1R2, a visualisation tool, to analyse the field at a macro level in terms of hotspot distribution, spatial and temporal distribution, and evolutionary trends, respectively. Next, we become specific and separate the prediction techniques for concrete compressive strength into two groups","PeriodicalId":510160,"journal":{"name":"International Journal of Advanced Research in Science, Communication and Technology","volume":"36 7","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Advanced Research in Science, Communication and Technology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48175/ijarsct-18237","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
When it comes to estimating, classifying, and forecasting material strength based on changing material parameters, machine learning (ML) techniques have shown to be dependable methodologies. It is found that choosing the right machine learning technique depends on the characteristics of the problem and the available data. Therefore, fifteen different machine learning techniques were used to a specific dataset of concrete compressive strength in order to assess the accuracy of ML models to predict concrete compressive strength. Due to its excellent performance while dealing with continuous target variables and nonlinear interactions among the features and the target, the Support Vector Regressor (SVR) had the greatest prediction accuracy (88.18%) of all the ML methods employed. To guarantee the structural integrity of building projects, it is essential to predict the breaking strength of concrete. The goal of this project is to create a machine learning model that can forecast concrete's breaking strength depending on the mix's composition and curing circumstances. A dataset was created that included details regarding concrete samples, such as mix ratios, curing temperatures, curing times, and breaking strengths. recise estimation of concrete's compressive strength is crucial for the advancement and construction. A bibliometric analysis of the pertinent literature published in was conducted in order to comprehend the state of research in the field of concrete compressive strength prediction. The previous ten years have seen the first research in this sector. The database consisted of 31,35 journal articles published between 2012 and 2021 in the Web of Science core database. The knowledge map was created using Cite Space 6.1R2, a visualisation tool, to analyse the field at a macro level in terms of hotspot distribution, spatial and temporal distribution, and evolutionary trends, respectively. Next, we become specific and separate the prediction techniques for concrete compressive strength into two groups
在根据不断变化的材料参数对材料强度进行估算、分类和预测时,机器学习(ML)技术已被证明是一种可靠的方法。研究发现,选择正确的机器学习技术取决于问题的特征和可用数据。因此,我们将 15 种不同的机器学习技术用于特定的混凝土抗压强度数据集,以评估 ML 模型预测混凝土抗压强度的准确性。由于支持向量回归器(SVR)在处理连续目标变量以及特征与目标之间的非线性交互时表现出色,因此在所有采用的 ML 方法中,SVR 的预测准确率最高(88.18%)。为保证建筑项目的结构完整性,预测混凝土的断裂强度至关重要。本项目的目标是创建一个机器学习模型,该模型可根据混合物的成分和养护情况预测混凝土的断裂强度。创建的数据集包括混凝土样本的详细信息,如混合比、养护温度、养护时间和断裂强度。为了了解混凝土抗压强度预测领域的研究现状,我们对 2010 年出版的相关文献进行了文献计量分析。在过去的十年中,该领域的研究刚刚起步。数据库包括 2012 年至 2021 年期间在 Web of Science 核心数据库中发表的 3135 篇期刊论文。我们使用可视化工具 Cite Space 6.1R2 绘制了知识地图,分别从热点分布、时空分布和演变趋势等方面对该领域进行了宏观分析。接下来,我们将混凝土抗压强度的预测技术具体分为两组