Smart Monitoring Kadar Formalin Berbasis Artificial Neural Network untuk Deteksi Keamanan Pangan

Fatoni, M. Imam (2024) Smart Monitoring Kadar Formalin Berbasis Artificial Neural Network untuk Deteksi Keamanan Pangan. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 2042211007-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (6MB) | Request a copy

Abstract

Penggunaan formalin sebagai pengawet makanan pada bahan pangan seperti ikan segar, ikan asin, tahu, tempe, dan ayam telah menjadi permasalahan serius karena dapat membahayakan kesehatan manusia. Oleh karena itu, diperlukan perangkat yang praktis, sensitif, dan akurat untuk mendeteksi keberadaan formalin pada bahan pangan. Penelitian ini mengembangkan alat pengukur kandungan formalin berbasis Artificial Neural Network (ANN) yang dilengkapi dengan sensor Grove HCHO, DHT-22, TGS-822 untuk mendeteksi formalin dan sensor RTD PT-100 untuk monitoring temperatur. Sistem pada alat dirancang untuk mendeteksi formalin pada berbagai variasi konsentrasi yang diaplikasikan pada tiga sampel bahan pangan, yaitu ikan segar, ikan asin, dan tahu. Pengujian menunjukkan bahwa alat telah mampu menjaga suhu pemanas pada set point 70°C secara otomatis dengan akurasi tinggi. Sensor yang digunakan juga menunjukkan kinerja optimal, dengan tingkat akurasi masing-masing sebesar 98,63% (Grove HCHO), 95,95% (TGS-822), 98,66% (DHT-22), dan 95,94% (RTD PT-100). Optimasi ANN menghasilkan hyperparameter terbaik yang meningkatkan akurasi prediksi hingga 95,31% pada ikan asin, 93,89% pada ikan segar, dan 88,78% pada tahu. Hasil penelitian tersebut membuktikan bahwa alat berbasis ANN efektif dan akurat dalam mendeteksi kandungan formalin pada bahan pangan. Dengan keunggulan tersebut, perangkat ini diharapkan dapat menjadi solusi praktis untuk meningkatkan pengawasan keamanan pangan serta mengurangi risiko penggunaan bahan tambahan berbahaya seperti formalin.
==============================================================================================================================
The use of formalin as a food preservative in products such as fresh fish, salted fish, tofu, tempeh, and chicken has become a serious issue due to its potential health hazards. Therefore, a practical, sensitive, and accurate device is needed to detect the presence of formalin in food products. This study developed a formalin content measurement device based on an Artificial Neural Network (ANN), equipped with Grove HCHO, DHT-22, and TGS-822 sensors for detecting formalin, and an RTD PT-100 sensor for temperature monitoring. The system is designed to detect formalin at various concentrations applied to three food samples: fresh fish, salted fish, and tofu. Testing demonstrated that the device effectively maintained the heater temperature at the set point of 70°C with high accuracy. The sensors used also showed optimal performance, with accuracy levels of 98.63% (Grove HCHO), 95.95% (TGS-822), 98.66% (DHT-22), and 95.94% (RTD PT-100). ANN optimization resulted in the best hyperparameters, improving prediction accuracy to 95,31% for salted fish, 93,89% for fresh fish, and 88.78% for tofu. These findings confirm that the ANN-based device is effective and accurate in detecting formalin content in food products. With these advantages, this device is expected to serve as a practical solution for enhancing food safety monitoring and reducing the risks associated with hazardous additives such as formalin.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Artificial Neural Network, Formalin, Keamanan bahan pangan, Formaldehyde, Food Safety
Subjects:	T Technology > T Technology (General) T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models.
Divisions:	Faculty of Vocational > Instrumentation Engineering
Depositing User:	M. Imam Fatoni
Date Deposited:	01 Feb 2025 23:19
Last Modified:	01 Feb 2025 23:19
URI:	http://repository.its.ac.id/id/eprint/117503

Actions (login required)

View Item