Identifikasi Karakter Kopi Arabika Berdasarkan Tingkat Roasting Menggunakan Array Sensor Gas Dengan Metode Fuzzy Mamdani

Najmilhana, Sukroeni (2021) Identifikasi Karakter Kopi Arabika Berdasarkan Tingkat Roasting Menggunakan Array Sensor Gas Dengan Metode Fuzzy Mamdani. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text Cover.pdf Download (698kB)
	Text Abstract.pdf Download (818kB)
	Text Abstrak.pdf Download (862kB)
	Text BAB 1.pdf Download (872kB)
	Text BAB 2.pdf Download (1MB)
	Text BAB 3.pdf Download (1MB)
	Text BAB 4.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text BAB 5.pdf Download (867kB)
	Text Daftar Pustaka.pdf Download (906kB)
	Text Lampiran.pdf Download (2MB)

Abstract

Coffee is one of the local commodity plants in Mount Slamet. One of them is Mount Malang Arabica coffee which is located on the slopes of Mount Slamet. Each type of coffee contains several volatile compounds that can affect the characteristics of the coffee itself. The aroma of coffee from the past until now is still characterized by involving several human testers with their sense of smell. This method uses quality control which is very subjective because it depends on the understanding, experience and ability of each coffee tester. This is considered less biased as a reference in classifying and identifying types of coffee based on their aroma. Therefore, a more alternative way is needed in activating the type of coffee based on its aroma, one of which is an electronic nose/gas sensor array. The purpose of this study was to determine the response pattern of the gas sensor array to the aroma of Mount Slamet Arabica coffee, the effect of roasting temperature and performance on the gas sensor array. The results of this study are expected to be a more alternative way of detecting the aroma of coffee beans and coffee grounds using an electronic nose/gas sensor array. The gas sensors used are MQ2, MQ3, MQ6, MQ7, and MQ135. The accuracy of the tool in identifying coffee is the percentage of success obtained is 89.286% and the error percentage is 10.714% from 28 tests. And the average gas sensor Array for training data is 256,509 for Light, 272,127 for medium, and 307,968 for Dark. From these results, a Mamdani fuzzy logic system was created in MATLAB software. Fuzzy testing in MATLAB can help to make decisions in identifying Mount Slamet coffee for Light, Medium, and Dark profiles. Keywords: Coffee scent, E-Nose, Gas Sensor, Fuzzy Logic

Actions (login required)

View Item