Varisa, Rahmawati (2024) Perancangan Spektrometer Pentakromatik Sederhana dengan Jaringan Syaraf Tiruan (JST) untuk Identifikasi Molekul Organik/Anorganik pada Air Limbah. Masters thesis, Institut Teknologi Sepuluh Nopember.
Text
6009212001-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 July 2026. Download (7MB) | Request a copy |
Abstract
Pencemaran lingkungan dikarenakan pembuangan air limbah yang tidak tertangani dengan baik dapat membahayakan kesehatan dan merusak lingkungan. Dalam penanganan air limbah, langkah identifikasi molekul polutan penting untuk menentukan solusi pengolahan air limbah yang tepat. Salah satu metode identifikasi molekul paling sederhana adalah dengan teknik spektroskopi menggunakan spektrometer. Spektrometer pentakromatik merupakan alat untuk menganalisis spektrum cahaya atau radiasi elektromagnetik dengan memisahkan cahaya menjadi lima warna utama. Pada penelitian ini, dirancang spektrometer dengan sumber cahaya pentakromatik serta memodelkan algoritma kecerdasan buatan untuk identifikasi molekul organik dan anorganik pada air limbah. Spektrometer pentakromatik difabrikasi secara sederhana untuk mengidentifikasi molekul organik dan anorganik berdasarkan karakteristik absorpsinya dari lima kombinasi warna LED (Light Emitting Diode) dengan panjang gelombang (warna): 395 nm (ungu), 442 nm (biru), 519 nm (hijau), 564 nm (kuning), dan 680 nm (merah). Fotodioda pada spektrometer mendeteksi cahaya melalui interaksi dengan material, di mana cahaya memicu eksitasi elektron, membentuk pasangan elektron dan lubang. Medan listrik internal pada fotodioda memisahkan pasangan ini, menghasilkan arus listrik yang dapat diukur. Molekul polutan diidentifikasi menggunakan model Jaringan Syaraf Tiruan (JST) dengan pendekatan Backpropagation tipe Cascade Forward Neural Network (CFNN), berdasarkan database spektral yang mencakup 16 jenis molekul organik dan anorganik. Algoritma JST yang digunakan telah dilengkapi dengan spektrometer pentakromatik, mencapai akurasi sebesar 97% dan Mean Squared Error (MSE) sebesar 0.0051769 dalam mendeteksi senyawa organik/anorganik dalam air limbah.
===================================================================================================================================
Environmental pollution due to wastewater disposal that is not handled properly can endanger health and damage environment. In wastewater treatment, step of identifying pollutant molecules is important to determine right wastewater treatment solution. One of simplest molecular identification methods is spectroscopic techniques using a spectrometer. A pentachromatic spectrometer is a tool for analyzing the spectrum of light or electromagnetic radiation by separating light into five main colors. A spectrometer with a pentachromatic light source was designed and an artificial intelligence algorithm was designed to identify organic and inorganic molecules in wastewater. The pentachromatic spectrometer is simply fabricated to identify organic and inorganic molecules based on their absorption characteristics from five LED (Light Emitting Diode) color combinations with wavelengths (colors): 395 nm (purple), 442 nm (blue), 519 nm (green), 564 nm (yellow), and 680 nm (red). Photodiode in spectrometer detects light through interaction with material, where light triggers excitation of electrons, forming electron and hole pairs. The internal electric field in photodiode separates this pair, producing a measurable electric current. Pollutant molecules were identified using an Artificial Neural Network (ANN) model with a Cascade Forward Neural Network (CFNN) type Backpropagation approach, based on a spectral database that includes 16 types of organic and inorganic molecules. The ANN algorithm used is equipped with a pentachromatic spectrometer, achieving an accuracy of 97% and a Mean Squared Error (MSE) of 0.0051769 in detecting and quantifying organic/inorganic compounds in wastewater.
Item Type: | Thesis (Masters) |
---|---|
Uncontrolled Keywords: | Air limbah, Database, Spektrometer, Pentakromatik; Wastewater, Database, Spectrometer, Cascade Forward Neural Network, Pentachromatic, Mean Squared Error (MSE) |
Subjects: | T Technology > T Technology (General) T Technology > TD Environmental technology. Sanitary engineering |
Divisions: | Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30101-(S2) Master Thesis |
Depositing User: | Varisa Rahmawati |
Date Deposited: | 20 Feb 2024 07:51 |
Last Modified: | 20 Feb 2024 07:51 |
URI: | http://repository.its.ac.id/id/eprint/107476 |
Actions (login required)
View Item |