Nugroho, Happy (2016) Rancang Bangun Sistim Identifikasi Gas Menggunakan Sinar UV-IR dan Metode Jaringan Syaraf Tiruan. Masters thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
2212204002-Master-Theses.pdf Download (3MB) | Preview |
Abstract
Teknik otomatisasi untuk mengklasifikasi jenis gas diperlukan di beberapa
aplikasi seperti monitoring lingkungan, pendeteksian terhadap kebocoran
gas, diagnosa kesehatan, optimalisasi proses industri dan lainnya. Hal ini
memerlukan teknik yang bersifat robust dan adaptif yang mampu merespon
perubahan konsentrasi gas secara dinamis. Pada penelitian ini telah
dilakukan rancang bangun sistim yang digunakan untuk mengukur dan
mengenali jenis gas. Peralatan ini menggunakan metode spektrofotometri
yang dilengkapi dengan Jaringan Syaraf Tiruan. Prinsip kerja dari alat ini
adalah cahaya yang terpancar dari deret 26 Light Emitting Diode (LED)
dengan panjang gelombang yang berbeda mulai dari ultraviolet 385 nm
sampai infrared 1720 nm diarahkan ke sampel gas yang berada di dalam
kuvet melalui serat optik. Sebagian cahaya yang telah melewati sampel gas
akan diteruskan menuju fotodioda. Deret LED tersebut menghasilkan pola
serapan panjang gelombang yang spesifik dan unik untuk setiap gas yang
diujikan. Pola sinyal ini diolah Jaringan Syaraf Tiruan untuk diklasifikan
agar dapat mengenali jenis gas secara otomatis. Hasil percobaan
menunjukkan bahwa Deret LED dengan panjang gelombang yang berbeda
dan pengenal pola Jaringan Syaraf Tiruan dapat mengenali jenis gas dengan
tingkat kebenaran 60%.
=============================================================================
Automation engineering for classifying the types of gas in several
applications are required in several applications such as monitoring of
environment, detecting of the gas leakage, diagnosing of health, optimizing
of the industrial process and other purposes. These applications require a
robust and adaptif technique that can adapt to any changes of gas
concentration dynamically. This research has designed a system that is used
for measuring and identifying the type of gas. This device uses the method of
spectrophotometric and artificial neural network. The principle work of this
device is the light emitted over twenty six Light Emitting Diode Array
(LEDs) with various wavelengths starting from Ultraviolet light-385 nm to
Infrared light-1720 nm were absorbed by the sample gas inside the chamber
through fiber optics. The lights that has passed through the sample gas
would be received by the photodiode. Those twenty six LEDs array
produced its specific and unique absorbance for each gas. This absorbed
signals would have been processed by artificial neural network as input
signal learning and identification process. The recognition system in this
study has 60% success rate.
| Item Type: | Thesis (Masters) |
|---|---|
| Additional Information: | RTE 543.85 Nug r |
| Uncontrolled Keywords: | deret LED; jaringan syaraf tiruan dan spektrofotometri; artificial neural network; LED array and spectrophotometric |
| Subjects: | Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TK Electrical engineering. Electronics Nuclear engineering |
| Divisions: | Faculty of Industrial Technology > Electrical Engineering > 20101-(S2) Master Thesis |
| Depositing User: | Anis Wulandari |
| Date Deposited: | 08 Jun 2017 05:26 |
| Last Modified: | 27 Dec 2018 03:21 |
| URI: | http://repository.its.ac.id/id/eprint/41532 |
Actions (login required)
 |
View Item |