Identifikasi Kualitas Beras Berbasis Sensor Gas dan Metode Backpropagation Neural Network

Mubarok, Moh. Hanif (2025) Identifikasi Kualitas Beras Berbasis Sensor Gas dan Metode Backpropagation Neural Network. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6022211033-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (2MB) | Request a copy

Abstract

Beras adalah salah satu komoditas pangan utama. Pengukuran kualitas beras hingga saat ini masih dilakukan secara manual, yang rentan terhadap kesalahan akibat keterbatasan penglihatan manusia dan subjektiitas penguji. Setelah panen, beras terus mengalami proses respirasi yang dipicu oleh aktivitas mikroba, yang menyebabkan penurunan kualitas selama penyimpanan. Penelitian ini mengembangkan sistem hidung elektronik yang dapat mengidentifikasi kualitas beras. Sistem ini memanfaatkan sensor gas MQ2, MQ135, dan TGS2602. Kualitas beras dapat dinilai dengan menganalisis pola gas karbondioksida (CO2), oksigen (O2), dan volatile organic compounds (VOC) selama penyimpanan. Proses pembelajaran dilakukan untuk mengenali pola-pola tersebut menggunakan metode Backpropagation Neural Network (BPNN). Hasil eksperimen dan pengujian sensor menunjukkan bahwa metode ini dapat mengidentifikasi dengan akurat baik beras berkualitas baik maupun buruk dengan tingkat akurasi 97%. Alat sensor ini juga diuji pada sampel beras yang dikemas dengan Modified Atmosphere Packaging (MAP) yaitu dengan penambahan gas CO2, N2, vakum, dan kelembapan.
==================================================================================================================================
Rice is one of the primary food commodities. The quality measurement of rice is mostly still conducted manually, which is prone to errors due to human visual limitations and examiner subjectivity. Post-harvest rice continues to undergo respiration processes driven by microbial activity, leading to deterioration during storage. This study developed an electronic nose system capable of identifying rice quality. The system utilizes MQ2, MQ135, and TGS2602 gas sensors. The quality of rice can be assessed by analyzing the carbon dioxide (CO2), oxigen (O2), and volatile organic compounds (VOC) gas patterns during storage. A learning process was conducted to recognize these patterns using the Backpropagation Neural Network (BPNN). Experimental results and sensor testing indicated that this method can accurately identify both Baik and Buruk quality of rice with an accuracy of 97%. The sensor device was also tested on rice samples packaged with Modified Atmosphere Packaging (MAP), including the addition of CO2, N2, vacuum, and humidity.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	BPNN, kualitas beras, makanan, sensor gas, food, gas sensors, MAP, rice quality
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Moh. Hanif Mubarok
Date Deposited:	31 Jan 2025 07:29
Last Modified:	31 Jan 2025 07:29
URI:	http://repository.its.ac.id/id/eprint/117516

Actions (login required)

View Item