Artificial intelligence prediction model for swelling potential of soil and quicklime activated rice husk ash blend for sustainable construction
Artificial intelligence (AI) algorithms of adaptive neuro-fuzzy inference system or the adaptive network-based fuzzy inference system (ANFIS) has been deployed to predict the swelling potential (SP) of treated weak soil. The soil was treated with quicklime activated rice husk ash (QARHA) and the pre...
保存先:
| 主要な著者: | , , , , , |
|---|---|
| フォーマット: | 論文 |
| 言語: | English |
| 出版事項: |
Penerbit Universiti Kebangsaan Malaysia
2021
|
| オンライン･アクセス: | http://journalarticle.ukm.my/18938/1/07.pdf http://journalarticle.ukm.my/18938/ https://www.ukm.my/jkukm/volume-334-2021/ |
| タグ: |
タグ追加
タグなし, このレコードへの初めてのタグを付けませんか!
|
| 要約: | Artificial intelligence (AI) algorithms of adaptive neuro-fuzzy inference system or the adaptive network-based fuzzy inference system (ANFIS) has been deployed to predict the swelling potential (SP) of treated weak soil. The soil was treated with quicklime activated rice husk ash (QARHA) and the prediction efficiency was compared with the previous outcomes of this operation from literature. The need for effective utilization of construction materials to achieve sustainable designs and monitoring of the behavior of built environment is the motivation behind the deployment of artificial intelligence in geo-environmental research and field operations. The use of ANFIS is common in different fields of science and business to predict the best fits from several data points. The results of this modeling exercise conducted with 25 datasets from mixture experimental treatment of soft soil with QARHA has shown that ANFIS is a better tool compared to the individual algorithms of ANN and FL and even the other artificial intelligence tools like scheffe, ANOVA, regression and extreme vertices methods. With performance index of 88% and correlation of about 71% in the ANFIS testing and 17% and 99% respectively in the ANFIS training, ANFIS proved to be a more powerful tool in achieving a more sustainable material utilization in earthwork constructions, design and monitoring of geotechnical systems performance. |
|---|