Mix Desain Engineered Cementitious Composite (ECC) Dengan Menggunakan Artificial Neural Network (ANN)

Tuanakotta, Abraham (2018) Mix Desain Engineered Cementitious Composite (ECC) Dengan Menggunakan Artificial Neural Network (ANN). Masters thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
3112202001-Master_Thesis.pdf

Download (3MB) | Preview

Abstract

Engineered Cementitious Composite (ECC) adalahmateria komposit semen dengan distribusi serat pendek secara acak dengan volume sekitar 2%. ECC mempunyai keunggulan daktalitasnya. Pemanfaatan ECC di Indonesia masih terbatas dan belum ada standar mix desainnya di Indonesia. Mix desain ECC selama ini digunakan dengan banyak percobaan, dan hal ini tidak efektif. Artificial Neural Network (ANN) adalah salah satu program komputer yang dapat memantu menentukan mix desain secara efektif. Penelitian ini membuat mix desain ECC secra eksperimen, dengan komposisi : w/c : 0.25- 0.40 , serat Polyvinil Alcohol (PVA) : 0 - 0.02 , Fly Ash (FA) : 0 - 6 , dan Superplasticizer (SP) : 0 - 0.1 sebagai data input. Pengujian kuat tekan dan kuat tarik ECC pada umur28 hari sebagai data output. Performance terbaik adalah kuat tekan w/c 0.35 dan 0.40 yaitu nilai Mean Square Error (MSE) = 2.31 , nilai regresi = 0.99 , dan nilai error = 2.62 % , mix desin untuk kuat tekan = 25 - 60 MPa. ============================================================ Engineered Cementitious Composite (ECC) is a cement composite material with random distribution of short fibers at a volume of about 2%. ECC has the advantage of ductility. Research to the use of ECC in Indonesia is still limited and there is no standard mix of its design in Indonesia. Mix design ECC has been carried out with many experiment, and they are not effective. Artificial Neural Network (ANN)is one of the computer program that can help determine the mix design effectively. The research made the mix design of ECC experimental, with the composition water to cement ratio rnges (w/c) : 0.25 - 0.40, Polyvinil Alcohol (PVA) fiber with presentation volume of 0 - 0.02 , Fly Ash (FA) with percentage of 0 - 0.6 , and Superplasticizer (SP) : 0-0.1 as input data. Testing compressive strength and tensile strength of ECC at 28 days as output data, and test data was modeled with the ANN program. The best performance is the compressive strenth with w/c is = 0.35 - 0.40, the value of MSE, Regression, percentage of Error are 2.31 ; 0.99 and 2.62% respectively. It was applied when the mix design for the compressive strength is the range of 25 - 60 MPa

Item Type: Thesis (Masters)
Uncontrolled Keywords: Cementitious engineered composite, Kuat tekan,Kuat tarik, Artificial Neural Network, Mean Square Error
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
T Technology > TA Engineering (General). Civil engineering (General) > TA681 Concrete construction
Divisions: Faculty of Civil Engineering and Planning > Civil Engineering > (S2) Master Theses
Depositing User: ABRAHAM TUANAKOTTA
Date Deposited: 18 Apr 2018 02:25
Last Modified: 18 Apr 2018 02:25
URI: http://repository.its.ac.id/id/eprint/50882

Actions (login required)

View Item View Item