Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning Techniques
The static elastic modulus (<i>Ec</i>) and compressive strength (<i>fc</i>) are critical properties of concrete. When determining <i>Ec</i> and <i>fc</i>, concrete cores are collected and subjected to destructive tests. However, destructive tests requi...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-06-01
|
| Series: | Materials |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1944/13/13/2886 |
| _version_ | 1797563991190405120 |
|---|---|
| author | Jong Yil Park Sung-Han Sim Young Geun Yoon Tae Keun Oh |
| author_facet | Jong Yil Park Sung-Han Sim Young Geun Yoon Tae Keun Oh |
| author_sort | Jong Yil Park |
| collection | DOAJ |
| description | The static elastic modulus (<i>Ec</i>) and compressive strength (<i>fc</i>) are critical properties of concrete. When determining <i>Ec</i> and <i>fc</i>, concrete cores are collected and subjected to destructive tests. However, destructive tests require certain test permissions and large sample sizes. Hence, it is preferable to predict <i>Ec</i> using the dynamic elastic modulus (<i>Ed</i>), through nondestructive evaluations. A resonance frequency test performed according to ASTM C215-14 and a pressure wave (P-wave) measurement conducted according to ASTM C597M-16 are typically used to determine <i>Ed</i>. Recently, developments in transducers have enabled the measurement of a shear wave (S-wave) velocities in concrete. Although various equations have been proposed for estimating <i>Ec</i> and <i>fc</i> from <i>Ed</i>, their results deviate from experimental values. Thus, it is necessary to obtain a reliable <i>Ed</i> value for accurately predicting <i>Ec</i> and <i>fc</i>. In this study, <i>Ed</i> values were experimentally obtained from P-wave and S-wave velocities in the longitudinal and transverse modes; <i>Ec</i> and <i>fc</i> values were predicted using these <i>Ed</i> values through four machine learning (ML) methods: support vector machine, artificial neural networks, ensembles, and linear regression. Using ML, the prediction accuracy of <i>Ec</i> and <i>fc</i> was improved by 2.5–5% and 7–9%, respectively, compared with the accuracy obtained using classical or normal-regression equations. By combining ML methods, the accuracy of the predicted <i>Ec</i> and <i>fc</i> was improved by 0.5% and 1.5%, respectively, compared with the optimal single variable results. |
| first_indexed | 2024-03-10T18:51:02Z |
| format | Article |
| id | doaj.art-1f6d35023cb044938a0d891988638323 |
| institution | Directory Open Access Journal |
| issn | 1996-1944 |
| language | English |
| last_indexed | 2024-03-10T18:51:02Z |
| publishDate | 2020-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Materials |
| spelling | doaj.art-1f6d35023cb044938a0d8919886383232023-11-20T05:07:33ZengMDPI AGMaterials1996-19442020-06-011313288610.3390/ma13132886Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning TechniquesJong Yil Park0Sung-Han Sim1Young Geun Yoon2Tae Keun Oh3Department of Safety Engineering, Seoul National University of Science and Technology, Seoul 01811, KoreaSchool of Civil, Architectural Engineering and Landscape Architecture, Sungkyunkwan University, Suwon 16419, KoreaDepartment of Safety Engineering, Incheon National University, Incheon 22012, KoreaDepartment of Safety Engineering, Incheon National University, Incheon 22012, KoreaThe static elastic modulus (<i>Ec</i>) and compressive strength (<i>fc</i>) are critical properties of concrete. When determining <i>Ec</i> and <i>fc</i>, concrete cores are collected and subjected to destructive tests. However, destructive tests require certain test permissions and large sample sizes. Hence, it is preferable to predict <i>Ec</i> using the dynamic elastic modulus (<i>Ed</i>), through nondestructive evaluations. A resonance frequency test performed according to ASTM C215-14 and a pressure wave (P-wave) measurement conducted according to ASTM C597M-16 are typically used to determine <i>Ed</i>. Recently, developments in transducers have enabled the measurement of a shear wave (S-wave) velocities in concrete. Although various equations have been proposed for estimating <i>Ec</i> and <i>fc</i> from <i>Ed</i>, their results deviate from experimental values. Thus, it is necessary to obtain a reliable <i>Ed</i> value for accurately predicting <i>Ec</i> and <i>fc</i>. In this study, <i>Ed</i> values were experimentally obtained from P-wave and S-wave velocities in the longitudinal and transverse modes; <i>Ec</i> and <i>fc</i> values were predicted using these <i>Ed</i> values through four machine learning (ML) methods: support vector machine, artificial neural networks, ensembles, and linear regression. Using ML, the prediction accuracy of <i>Ec</i> and <i>fc</i> was improved by 2.5–5% and 7–9%, respectively, compared with the accuracy obtained using classical or normal-regression equations. By combining ML methods, the accuracy of the predicted <i>Ec</i> and <i>fc</i> was improved by 0.5% and 1.5%, respectively, compared with the optimal single variable results.https://www.mdpi.com/1996-1944/13/13/2886concretestatic elastic modulusdynamic elastic moduluscompressive strengthmachine learningP-wave |
| spellingShingle | Jong Yil Park Sung-Han Sim Young Geun Yoon Tae Keun Oh Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning Techniques Materials concrete static elastic modulus dynamic elastic modulus compressive strength machine learning P-wave |
| title | Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning Techniques |
| title_full | Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning Techniques |
| title_fullStr | Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning Techniques |
| title_full_unstemmed | Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning Techniques |
| title_short | Prediction of Static Modulus and Compressive Strength of Concrete from Dynamic Modulus Associated with Wave Velocity and Resonance Frequency Using Machine Learning Techniques |
| title_sort | prediction of static modulus and compressive strength of concrete from dynamic modulus associated with wave velocity and resonance frequency using machine learning techniques |
| topic | concrete static elastic modulus dynamic elastic modulus compressive strength machine learning P-wave |
| url | https://www.mdpi.com/1996-1944/13/13/2886 |
| work_keys_str_mv | AT jongyilpark predictionofstaticmodulusandcompressivestrengthofconcretefromdynamicmodulusassociatedwithwavevelocityandresonancefrequencyusingmachinelearningtechniques AT sunghansim predictionofstaticmodulusandcompressivestrengthofconcretefromdynamicmodulusassociatedwithwavevelocityandresonancefrequencyusingmachinelearningtechniques AT younggeunyoon predictionofstaticmodulusandcompressivestrengthofconcretefromdynamicmodulusassociatedwithwavevelocityandresonancefrequencyusingmachinelearningtechniques AT taekeunoh predictionofstaticmodulusandcompressivestrengthofconcretefromdynamicmodulusassociatedwithwavevelocityandresonancefrequencyusingmachinelearningtechniques |