Artificial Intelligence Approaches for Optimizing High-Performance Concrete Mix Design  

Oleh:

Prayogo, D. ; Cheng, M-Y. ; Wibowo, D. K.

Jenis: Article from Proceeding
Dalam koleksi: Proceeding The 1st International Conference on Sustainable Civil Engineering Structures and Construction Materials (SCESCM) di Yogyakarta, September 11 – 13, 2012, page 267-273.
Topik: High-Performance Concrete; Genetic Algorithm; Evolutionary Support Vector Machine.
Fulltext: 37.pdf (556.66KB)

Isi artikel

Traditional mix proportioning methods were not sufficient to deal with nonlinear relationship among components and high-performance concrete (HPC) properties due to the expensive cost, limitation of use, and incompetence. Therefore, finding the suitable technology to optimize the mix design of HPC might deliver significant benefits to construction industry. This research introduces artificial intelligence (AI) approaches to search for the optimum mixture composition of HPC and the lowest cost mixture which yield required strength. AI approaches involved in this research include Evolutionary Support Vector Machine Inference Model (ESIM) and K-means Chaos Genetic Algorithm (KCGA). ESIM is employed for mapping the complex relationship between composition and compressive strength of HPC. Meanwhile, KCGA is used to conduct optimization process for searching the optimum mixture composition. Total 1030 records from HPC laboratory experiments are provided to demonstrate the model application. According to the results, ESIM can build accurate models for predicting the compressive strength and KCGA can generate the lowest cost mixtures for the given required strength.

Opini Anda

Klik untuk menuliskan opini Anda tentang koleksi ini!

Kembali