Identifikasi Daging Segar terhadap Daging Busuk dengan Menggunakan Sensor Polimer Konduktif dan Jaring Saraf Tiruan (JST)

Abstract

Fresh or rotten meat is a different matter. Damage to the meat will produce a distorted odor, mucus, discoloration in certain areas and an undesirable taste due to the formation of metabolism. The odor is described as fishy, ​​rotten, containing sulfur and like ammonia. In this research, the author discusses a system for identifying the condition of meat based on the odor that arises from meat in three states, namely odorless odor, fresh odor and rotten odor. Fresh odor is taken from meat odor that is within 1 (one) day after being cut and rotten odor is taken from meat odor that is on the 2nd day. In this study, the test sample meat was placed in a closed container at room temperature for 2 days. Data was taken for 2 days from meat odors of known type. The sensor array consists of eight sensors made of conducting polymer material. The polymer materials used are silicon DC-200, PEG-20M, 0V-101, 0V-17, DEGA, PEG-200, PEG-1540, and PEG-6000 mixed with Carbon black. A two-layer artificial neural net consisting of eight input nodes and three output neurons, was trained using the Kohonen algorithm with a training process that was completed in 4 iterations. From 20 tests, 10 times exposure to steam from fresh odor and 10 times exposure to steam from rotten odor, carried out alternately, it was found that the system failed twice. Thus, the system success percentage reached 95 percent.