A data-driven technique to predict the compressive strength of sugarcane bagasse ash concrete

Nariman Saeed, Gholamreza Pazouki, Won Hee Kang, Zhong Tao

Research output: Book chapter/Published conference paperConference paperpeer-review

Abstract

Sugarcane bagasse ash (SCBA) is an agricultural and industrial waste material produced in large quantities each year. Traditionally used as fertilizer or buried, SCBA can also be incorporated into concrete to reduce cement consumption, mitigate environmental impact, and improve mechanical properties. This study explores the effectiveness of several AI models, including radial basis function neural network (RBFNN), artificial neural network (ANN), and support vector regression (SVR), in predicting the compressive strength of concrete with SCBA. A dataset of 1,819 data points from previous studies was used, categorised into three groups: 340 data points for concrete with SCBA, 459 for concrete without any SCMs, and 1,020 for concrete with other supplementary cementitious materials. The models were tested in three configurations: (1) trained only on SCBA data, (2) trained on SCBA data combined with concrete data without SCMs, and (3) trained on the entire dataset. Results showed that the SVR model performed best when trained solely on SCBA data, while the RBFNN model excelled when using the combined dataset, demonstrating improved accuracy after refining the non-SCM data.
Original languageEnglish
Title of host publicationThe 8th Asian-Pacific Symposium on Structural Reliability and Its Applications (APSSRA2024)
Publication statusPublished - 2024
Event8th Asian-Pacific Symposium on Structural Reliability and Its Applications 2024: APSSRA - Ulsan Exhibition and Convention Center, Ulsan, Korea, Democratic People's Republic of
Duration: 29 Sept 202402 Oct 2024
https://apssra2024.kr/
https://apssra2024.kr/download/Full%20program_0924_F.pdf (Program)

Conference

Conference8th Asian-Pacific Symposium on Structural Reliability and Its Applications 2024
Country/TerritoryKorea, Democratic People's Republic of
CityUlsan
Period29/09/2402/10/24
OtherThe first Asian-Pacific Symposium on Structural Reliability and Its Applications (APSSRA) was held in Tokyo, Japan, in November 1995. Since then, APSSRA has been held every four years between ICASP and ICOSSAR in various Asian cities including Taipei, Seoul, Hong Kong, Singapore, Shanghai, and Tokyo. The 8th APSSRA (APSSRA 2024) is set to take place in Ulsan, Republic of Korea, from 29 September to 2 October 2024. This year's symposium is organized by the Ulsan National Institute of Science and Technology (UNIST), which leads fusion research and technology development in Korea. Join us in Ulsan, one of the seven major cities of Korea and a seaport town, as we delve into the latest challenges and advances in structural reliability, resilience, and sustainable development. Your participation and scientific contributions will be key to the success of APSSRA 2024. We look forward to seeing you in Ulsan!
Internet address

Fingerprint

Dive into the research topics of 'A data-driven technique to predict the compressive strength of sugarcane bagasse ash concrete'. Together they form a unique fingerprint.

Cite this