CD Skripsi

Prakiraan Nilai Permeabilitas Berdasarkan Sifat Fisik Tanah Menggunakan Artificial Neural Network Berbasis Algoritma Backprogation

---

ABSTRACT
Permeability is a parameter of soil related to the most important thing in
development to understand infiltration, runoff, drainage and settlement processes.
Soil permeability testing can be carried out through laboratory or field testing
procedures, but there is no definite reference regarding the correlation of this
permeability with the parameters of soil physical properties and soil mechanical
properties. Artificial neural networks are often used to estimate complex and
nonlinear values. The aim of this research is to estimate the permeability coefficient
(k) based on permeability test data and soil physical properties in the laboratory
using the ability of artificial neural networks with the backpropagation algorithm.
This study was divided into 2 stages, for expansive and non-expansive soil types,
which were divided into 2 training variations, namely 80:20 and 70:30, with input
data in the form of soil liquid limit (LL), soil plasticity index (IP), and % fine grain.
This study shows that the artificial neural network is capable and effective in
predicting the value of the permeability coefficient (k) with a small error value and
a strong regression equation where R and R2 are close to 1 with small RMSE and
MAE. The best network structure obtained has 2 hidden layers with 40 neurons in
the first hidden layer and 20 neurons in the second hidden layer in both types of
soil. Meanwhile, with the help of a simple matrix, the appliaction of the network is
successful in determining the prediction of new input value which result in small
RMSE and MAE accruracy on expansive and non-expansive soils
Keyword: Permeability, Soil Physical Properties, Expansive Soil, Non-Expansive
Soil, Artificial Neural Network, Backpropagation.

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

1807113471

Penerbit

Pekanbaru : Universitas Riau – Fakultas Teknik – Teknik Sipil., 2023

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

1807113471

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