Artificial neural network approach using Levenberg-Marquardt algorithm in the use of bottom ash waste in concrete hollow block design

Date of Publication

2003

Document Type

Master's Thesis

Degree Name

Master of Science in Civil Engineering

College

Gokongwei College of Engineering

Department/Unit

Civil Engineering

Thesis Adviser

Andres Winston C. Oreta

Defense Panel Chair

Ronaldo S. Gallardo

Defense Panel Member

Alexis M. Fillone
Eugenio Chan

Abstract/Summary

Neural Network modeling was applied for the prediction of compressive strength of Coal Bottom Ash (CBA). Levenberg-Marquardt was used for the different neural network architectures to find acceptable models than can accurately predict the compressive strength of CHB's and realistically model the behavior of CHB's with CBA as partial substitute to sand. In addition, the maximum percentage of CBA content was derived from the ANN (Artificial Neural Network) model based on PNS (Philippine National Standards) types. CBA is a waste by-product of coal-fired power plant. An experimental study utilizing CBA as a partial substitute to sand in the production of CNB's was conducted. Around 429 pieces of four-inch thick CHB's were tested with such variable mix proportions as: water-cement ratio (w/c), cement-aggregate ratio (c/a), weight of the specimen (wt), slump (sl), and coal bottom ash percent substitution (CBA) taken into consideration.

Abstract Format

html

Language

English

Format

Print

Accession Number

TG03517

Shelf Location

Archives, The Learning Commons, 12F Henry Sy Sr. Hall

Physical Description

86 numb. leaves ; 28 cm.

Keywords

Neural networks (Computer science); Evolutionary computation; Algorithms; Waste products; Concrete blocks

This document is currently not available here.

Share

COinS