Prediction of Unconfined Compressive Strength in Stabilized Clay Soil Using Artificial Neural Networks
Expansive clay is a problematic type of soil because it has large shrinkage properties. One action that can be taken to improve problematic soil is to stabilize it with additives such as lime, cement, RHA, fly ash, and GGBS. The results of stabilization using additives like this can increase the str...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2024-01-01
|
| Series: | BIO Web of Conferences |
| Subjects: | |
| Online Access: | https://www.bio-conferences.org/articles/bioconf/pdf/2024/63/bioconf_sage-grace2024_06002.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850245505372651520 |
|---|---|
| author | Indriani Lia Riyadi Slamet Zaki Ahmad |
| author_facet | Indriani Lia Riyadi Slamet Zaki Ahmad |
| author_sort | Indriani Lia |
| collection | DOAJ |
| description | Expansive clay is a problematic type of soil because it has large shrinkage properties. One action that can be taken to improve problematic soil is to stabilize it with additives such as lime, cement, RHA, fly ash, and GGBS. The results of stabilization using additives like this can increase the strength value of clay soil. Artificial Neural Networks (ANN) have been introduced in the geotechnical field to predict different soil properties. This research develops an artificial neural networks model to predict the Unconfined Compressive Strength (UCS) value of soil that has been stabilized, this is because the artificial neural networks model can show superior prediction results due to its flexibility and adaptability in generating data. The amount of data in this test was 420 and was divided into 336 training data and 84 testing data. In carrying out the training phase, 13 inputs were used in the form of granulometric test results, and in the testing phase, data from soil-free compression tests in the laboratory were used. The result of this research is that the use of the artificial neural networks model can predict the soil unconfined compressive strength value accurately because it gets a coefficient of determination value of 0.99229 which is almost close to number one. |
| format | Article |
| id | doaj-art-e42fad6d1be54d7b95cb5f27051cbb35 |
| institution | OA Journals |
| issn | 2117-4458 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | EDP Sciences |
| record_format | Article |
| series | BIO Web of Conferences |
| spelling | doaj-art-e42fad6d1be54d7b95cb5f27051cbb352025-08-20T01:59:26ZengEDP SciencesBIO Web of Conferences2117-44582024-01-011440600210.1051/bioconf/202414406002bioconf_sage-grace2024_06002Prediction of Unconfined Compressive Strength in Stabilized Clay Soil Using Artificial Neural NetworksIndriani Lia0Riyadi Slamet1Zaki Ahmad2Magister of Civil Engineering, University Muhammadiyah YogyakartaDepartment of Information Technology, University Muhammadiyah YogyakartaMagister of Civil Engineering, University Muhammadiyah YogyakartaExpansive clay is a problematic type of soil because it has large shrinkage properties. One action that can be taken to improve problematic soil is to stabilize it with additives such as lime, cement, RHA, fly ash, and GGBS. The results of stabilization using additives like this can increase the strength value of clay soil. Artificial Neural Networks (ANN) have been introduced in the geotechnical field to predict different soil properties. This research develops an artificial neural networks model to predict the Unconfined Compressive Strength (UCS) value of soil that has been stabilized, this is because the artificial neural networks model can show superior prediction results due to its flexibility and adaptability in generating data. The amount of data in this test was 420 and was divided into 336 training data and 84 testing data. In carrying out the training phase, 13 inputs were used in the form of granulometric test results, and in the testing phase, data from soil-free compression tests in the laboratory were used. The result of this research is that the use of the artificial neural networks model can predict the soil unconfined compressive strength value accurately because it gets a coefficient of determination value of 0.99229 which is almost close to number one.https://www.bio-conferences.org/articles/bioconf/pdf/2024/63/bioconf_sage-grace2024_06002.pdfunconfined compressive strengthpredictionartificial neural networksstabilization |
| spellingShingle | Indriani Lia Riyadi Slamet Zaki Ahmad Prediction of Unconfined Compressive Strength in Stabilized Clay Soil Using Artificial Neural Networks BIO Web of Conferences unconfined compressive strength prediction artificial neural networks stabilization |
| title | Prediction of Unconfined Compressive Strength in Stabilized Clay Soil Using Artificial Neural Networks |
| title_full | Prediction of Unconfined Compressive Strength in Stabilized Clay Soil Using Artificial Neural Networks |
| title_fullStr | Prediction of Unconfined Compressive Strength in Stabilized Clay Soil Using Artificial Neural Networks |
| title_full_unstemmed | Prediction of Unconfined Compressive Strength in Stabilized Clay Soil Using Artificial Neural Networks |
| title_short | Prediction of Unconfined Compressive Strength in Stabilized Clay Soil Using Artificial Neural Networks |
| title_sort | prediction of unconfined compressive strength in stabilized clay soil using artificial neural networks |
| topic | unconfined compressive strength prediction artificial neural networks stabilization |
| url | https://www.bio-conferences.org/articles/bioconf/pdf/2024/63/bioconf_sage-grace2024_06002.pdf |
| work_keys_str_mv | AT indrianilia predictionofunconfinedcompressivestrengthinstabilizedclaysoilusingartificialneuralnetworks AT riyadislamet predictionofunconfinedcompressivestrengthinstabilizedclaysoilusingartificialneuralnetworks AT zakiahmad predictionofunconfinedcompressivestrengthinstabilizedclaysoilusingartificialneuralnetworks |