Klasifikasi Kualitas Daging Ayam Berbasis Gas Sensor Array dan Probabilistic Neural Network pada Raspberry PI

Azradenis, Muhamad Dzaki (2026) Klasifikasi Kualitas Daging Ayam Berbasis Gas Sensor Array dan Probabilistic Neural Network pada Raspberry PI. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5022221067-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Daging ayam merupakan sumber protein hewani yang banyak dikonsumsi, tetapi mudah mengalami penurunan kualitas pada suhu ruang. Aktivitas mikroorganisme dan dekomposisi protein menghasilkan senyawa volatil, seperti amonia (NH₃), hidrogen sulfida (H₂S), dan senyawa organik volatil lainnya. Pada tahap awal pembusukan, perubahan visual dan aroma belum selalu terdeteksi dengan jelas sehingga penilaian organoleptik cenderung subjektif. Tugas akhir ini merancang sistem klasifikasi kualitas daging ayam menggunakan larik sensor gas dan metode Probabilistic Neural Network (PNN) yang dijalankan secara lokal pada Raspberry Pi. Sensor MQ-135, MQ-136, dan MQ-137 digunakan untuk mendeteksi gas volatil dari sampel daging ayam di dalam chamber. ESP32 mengakuisisi data analog melalui analog-to-digital converter (ADC) dan mengirimkannya ke Raspberry Pi melalui komunikasi serial. Model PNN mengklasifikasikan sampel ke dalam tiga kelas operasional, yaitu Segar, Mulai Membusuk, dan Busuk. Ketiga kelas tersebut ditentukan berdasarkan interval waktu penyimpanan pada suhu ruang, bukan berdasarkan hasil pengujian mikrobiologi. Sistem diuji menggunakan 27.030 baris data yang terdistribusi secara seimbang pada ketiga kelas. Evaluasi menggunakan confusion matrix menghasilkan akurasi keseluruhan sebesar 83,56%. Kinerja terbaik diperoleh pada kelas Busuk, sedangkan kesalahan paling banyak terjadi antara kelas Segar dan Mulai Membusuk akibat pola respons sensor yang masih beririsan pada fase transisi. Secara keseluruhan, sistem mampu melakukan klasifikasi secara lokal dan waktu nyata, meskipun kestabilan sensor dan kondisi chamber masih perlu ditingkatkan.
======================================================================================================================================
Chicken meat is widely consumed as a source of animal protein but deteriorates rapidly at room temperature. Microbial activity and protein decomposition produce volatile compounds, including ammonia (NH₃), hydrogen sulfide (H₂S), and other volatile organic compounds. During early spoilage, changes in appearance and odor may not be clearly detected, making organoleptic assessment subjective. This study developed a chicken meat quality classification system using a gas sensor array and a Probabilistic Neural Network (PNN) implemented locally on a Raspberry Pi. MQ-135, MQ-136, and MQ-137 sensors were used to detect volatile gases emitted by chicken meat samples inside a chamber. An ESP32 acquired analog sensor data through an analog-to-digital converter (ADC) and transmitted the data to the Raspberry Pi via serial communication. The PNN model classified the samples into three operational classes: Fresh, Early Spoilage, and Spoiled. These classes were determined based on room-temperature storage intervals rather than microbiological test results. The system was evaluated using 27,030 data records equally distributed among the three classes. Confusion matrix evaluation showed an overall accuracy of 83.56%. The highest performance was achieved for the Spoiled class, while most errors occurred between the Fresh and Early Spoilage classes due to overlapping sensor responses during the transition stage. Overall, the system performed local and real-time classification, although sensor stability and chamber conditions require further improvement.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	daging ayam, larik sensor gas, Probabilistic Neural Network, Raspberry Pi, klasifikasi kualitas, Chicken meat, gas sensor array, Probabilistic Neural Network, Raspberry Pi, quality classification.
Subjects:	T Technology > TA Engineering (General). Civil engineering (General) > TA1573 Detectors. Sensors T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Muhamad Dzaki Azradenis
Date Deposited:	28 Jul 2026 02:25
Last Modified:	28 Jul 2026 02:52
URI:	http://repository.its.ac.id/id/eprint/138810

Actions (login required)

View Item