The Optimal Machine Learning Model for the Precise Prediction of HighPerformance Concrete Strength Property
The prominent mechanical property of concrete is compressive strength which guarantees the performance and safety of the structure in its life cycle. Assessment of compressive strength, especially for high-performance concrete, is first due to the nonlinear relationship between the compressive stren...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Bilijipub publisher
2023-03-01
|
| Series: | Advances in Engineering and Intelligence Systems |
| Subjects: | |
| Online Access: | https://aeis.bilijipub.com/article_169080_e5c89ef6789706e42ce394f91f2af5bb.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823856398635630592 |
|---|---|
| author | Yufeng Qian |
| author_facet | Yufeng Qian |
| author_sort | Yufeng Qian |
| collection | DOAJ |
| description | The prominent mechanical property of concrete is compressive strength which guarantees the performance and safety of the structure in its life cycle. Assessment of compressive strength, especially for high-performance concrete, is first due to the nonlinear relationship between the compressive strength and the concrete constituents. Second, the supplementary cementitious materials admixed with the mix design of high-performance concrete encounter difficulties. The machine learning-based method, which relies on data mining, helps develop reliable and precise models to predict compressive strength. The present study employs a machine learning-based support vector regression (SVR) method to implement compressive strength prediction. The model accuracy was enhanced and strengthened by tuning the practical constraints of the support vector regression method. Marine predator and grasshopper optimization algorithms are performing the tuning process. The results of hybrid models show that the marine predator-based algorithm (MPA-SVR) played better than the grasshopper-based model (GOA-SVR) in predicting the compressive strength of high-performance concrete. The values of R2 for MPA-SVR and GOA-SVR are reported as 0.9939 and 0.9873, which implies that the MPA-SVR is more capable of implementing the compressive strength prediction. |
| format | Article |
| id | doaj-art-0ebb76d9de1e4ac7b930906779eab248 |
| institution | Kabale University |
| issn | 2821-0263 |
| language | English |
| publishDate | 2023-03-01 |
| publisher | Bilijipub publisher |
| record_format | Article |
| series | Advances in Engineering and Intelligence Systems |
| spelling | doaj-art-0ebb76d9de1e4ac7b930906779eab2482025-02-12T08:47:02ZengBilijipub publisherAdvances in Engineering and Intelligence Systems2821-02632023-03-0100201506110.22034/aeis.2023.381503.1065169080The Optimal Machine Learning Model for the Precise Prediction of HighPerformance Concrete Strength PropertyYufeng Qian0School of Science, Hubei University of Technology, Wuhan, 430068, China.The prominent mechanical property of concrete is compressive strength which guarantees the performance and safety of the structure in its life cycle. Assessment of compressive strength, especially for high-performance concrete, is first due to the nonlinear relationship between the compressive strength and the concrete constituents. Second, the supplementary cementitious materials admixed with the mix design of high-performance concrete encounter difficulties. The machine learning-based method, which relies on data mining, helps develop reliable and precise models to predict compressive strength. The present study employs a machine learning-based support vector regression (SVR) method to implement compressive strength prediction. The model accuracy was enhanced and strengthened by tuning the practical constraints of the support vector regression method. Marine predator and grasshopper optimization algorithms are performing the tuning process. The results of hybrid models show that the marine predator-based algorithm (MPA-SVR) played better than the grasshopper-based model (GOA-SVR) in predicting the compressive strength of high-performance concrete. The values of R2 for MPA-SVR and GOA-SVR are reported as 0.9939 and 0.9873, which implies that the MPA-SVR is more capable of implementing the compressive strength prediction.https://aeis.bilijipub.com/article_169080_e5c89ef6789706e42ce394f91f2af5bb.pdfcompressive strengthhigh-performance concretemarine predator algorithmgrasshopper optimization algorithmsupport vector regression |
| spellingShingle | Yufeng Qian The Optimal Machine Learning Model for the Precise Prediction of HighPerformance Concrete Strength Property Advances in Engineering and Intelligence Systems compressive strength high-performance concrete marine predator algorithm grasshopper optimization algorithm support vector regression |
| title | The Optimal Machine Learning Model for the Precise Prediction of HighPerformance Concrete Strength Property |
| title_full | The Optimal Machine Learning Model for the Precise Prediction of HighPerformance Concrete Strength Property |
| title_fullStr | The Optimal Machine Learning Model for the Precise Prediction of HighPerformance Concrete Strength Property |
| title_full_unstemmed | The Optimal Machine Learning Model for the Precise Prediction of HighPerformance Concrete Strength Property |
| title_short | The Optimal Machine Learning Model for the Precise Prediction of HighPerformance Concrete Strength Property |
| title_sort | optimal machine learning model for the precise prediction of highperformance concrete strength property |
| topic | compressive strength high-performance concrete marine predator algorithm grasshopper optimization algorithm support vector regression |
| url | https://aeis.bilijipub.com/article_169080_e5c89ef6789706e42ce394f91f2af5bb.pdf |
| work_keys_str_mv | AT yufengqian theoptimalmachinelearningmodelfortheprecisepredictionofhighperformanceconcretestrengthproperty AT yufengqian optimalmachinelearningmodelfortheprecisepredictionofhighperformanceconcretestrengthproperty |