Sweetness Analyzer Polikromatik Sederhana dengan Artificial Neural Network untuk Identifikasi Jenis Pemanis dalam Larutan

Prasetyo, Andry (2024) Sweetness Analyzer Polikromatik Sederhana dengan Artificial Neural Network untuk Identifikasi Jenis Pemanis dalam Larutan. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5001201036-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 July 2026. Download (3MB) | Request a copy

Abstract

Konsumsi pemanis buatan secara terus-menerus dapat meningkatkan risiko diabetes tipe 2, obesitas, penyakit kardiovaskular, kanker, karies gigi, bahkan kematian pada orang dewasa. Dalam penanganan risiko ini penting untuk mengetahui jenis pemanis pada minuman. Metode indentifikasi kandungan gula dan pemanis buatan saat ini terdapat teknik high-performance liquid chromatography (HPLC) dan spektroskopi FTIR (Fourier Transform Infrared). Sayangnya kedua metode tersebut memerlukan keahlian khusus dan memerlukan waktu yang relatif lama dalam proses identifikasi. Pada tugas akhir ini, Sweetness Analyzer polikromatik sederhana telah difabrikasi untuk mampu mengidentifikasi kandungan gula dan pemanis buatan berdasarkan karakteristik warna, pH, dan absorpsinya dari sensor TCS3200, sensor pH, dan fotodioda. Dengan memanfaatkan lampu pijar, alat mampu bekerja dengan cahaya polikromatik. Identifikasi kandungan gula dan pemanis buatan dilakukan Artificial Neural Network (ANN) dengan pendekatan backpropagation tipe Feed Forwad Neural Network (FFNN) berdasarkan database RGB, pH, dan absorpsi model kandungan gula dan pemanis buatan. Algoritma ANN-FFNN menghasilkan tingkat keakuratan identifikasi kandungan terbaik mencapai 83.3333% dengan Mean Square Error (MSE) sebesar 0,0258. Database Sweetness Analyzer Polikromatik Sederhana telah divalidasi dengan refractometer konvensional. Sweetness Analyzer Polikromatik Sederhana juga telah diuji untuk mengidentifikasi jenis pemanis larutan homogen, heterogen, dan minuman teh kemasan. Tetapi berdasarkan uji coba belum mampu mengidentifikasi kandungan gula dan pemanis buatan dalam minuman yang diluar jangkauan database. Oleh karena itu, performa Sweetness Analyzer polikromatik sederhana dapat ditingkatkan dengan memperluas kapasitas identifikasi yaitu dengan memperkaya database RGB, pH, dan absorbansi gula dari berbagai jenis pemanis.
==================================================================================================================================
Continuous consumption of artificial sweeteners can increase the risk of type 2 diabetes, obesity, cardiovascular diseases, cancer, dental caries, and even mortality in adults. In managing this risk, it is important to determine the types of sweeteners in beverages. Currently, methods for identifying sugar and artificial sweetener content include high-performance liquid chromatography (HPLC) and Fourier Transform Infrared spectroscopy (FTIR). Unfortunately, both methods require specialized skills and relatively long identification processes. In this final project, a simple polychromatic Sweetness Analyzer has been fabricated to identify sugar and artificial sweetener content based on color characteristics, pH, and absorption using the TCS3200 sensor, pH sensor, and photodiode. By utilizing an incandescent lamp, the device can work with polychromatic light. Identification of sugar and artificial sweetener content is done using an Artificial Neural Network (ANN) with a Feed Forward Neural Network (FFNN) backpropagation approach based on RGB, pH, and absorption database models of sugar and artificial sweeteners. The ANN-FFNN algorithm achieves the best identification accuracy level of 83.3333% with a Mean Square Error (MSE) of 0.0258. The database of the Simple Polychromatic Sweetness Analyzer has been validated with a conventional refractometer. The Simple Polychromatic Sweetness Analyzer has also been tested to identify the types of sweeteners in homogeneous, heterogeneous solutions, and packaged tea beverages. However, based on the tests, it has not been able to identify sugar and artificial sweetener content in beverages beyond the scope of the database. Therefore, the performance of the Simple Polychromatic Sweetness Analyzer can be enhanced by expanding its identification capacity, namely by enriching the RGB, pH, and sugar absorption database with various types of sweeteners.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Sweetness Analyzer, Database, FFNN, Larutan, Polikromatik; Sweetness Analyzer, Database, FFNN, Solution, Polychromatic.
Subjects:	Q Science > QC Physics > QC451 Spectroscopy
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Physics > 45201-(S1) Undergraduate Thesis
Depositing User:	Andry Prasetyo
Date Deposited:	12 Feb 2024 02:58
Last Modified:	12 Feb 2024 02:58
URI:	http://repository.its.ac.id/id/eprint/106845

Actions (login required)

View Item