CD Skripsi

Prediksi Kuat Tekan Bata Ringan Pasca Paparan Panas Dengan Metode Jaringan Saraf Tiruan

---

ABSTRACT
The purpose of this study was to examine the application of artificial neural
networks to predict the compressive strength of lightweight concrete after
exposure to heat, and density before and after exposure to heat. To achieve this
goal, an experiment was carried out by testing ligthweight concrete as many as
144 samples. The dimensions of the sample are 10x10x10 cm. Before being
exposed to heat, the sampels were stored at room temperature for 28 days and
tested for density. The number of samples are 24 samples for each variation of
heat exposure temperature. The variations of the heat exposure temperature used
were respectively 100ºC, 150ºC, 200ºC, 250ºC, 300ºC, and room temperature.
This heat exposure is done by putting the sample into the oven for two hours. Then
the sample is removed and cooled at room temperature. Furthermore, density and
compressive strength tests were carried out after exposure to heat. For numerical
modeling of compressive strength, after being exposed to heat, an analysis was
carried out using an artificial neural network. Analysis using an artificial neural
network consists of two stages, which are the training stage and the validation
stage. At the training stage, the data from the test results of compressive strength,
density, temperature of heat exposure and thermal conductivity are used as input.
Furthermore, the results of the training are used at the validation stage. Based on
the analysis of artificial neural networks that utilize feed forward
backpropagation architecture and the Levenberg-Marquardt training algorithm,
the most optimal analysis is obtained on 6 hidden layer neurons which produces a
Mean Squared Error of 0.00841 and a Regression of 0.95584. Based on the
results of Mean Squared Error which is close to 0 and Regression is close to 1, it
indicates that the artificial neural network can accurately predict the compressive
strength of lightweight bricks after heat exposure.
Keywords: Artificial neural network, compressive strength, lightweight concrete,
heat exposure, prediction

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

07 01. 122 (0028)

Penerbit

Pekanbaru : Universitas Riau –Fakultas Teknik – Teknik Sipil., 2022

Deskripsi Fisik

ix, 79 hlm.; ill.: 29 cm

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

07 01. 122 (0028)

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

TEKNIK SIPIL

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Daus

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar