Implementasi Sensor Gas Dan Suara Pada Pengendalian Proses Sangrai Kopi Menggunakan Fuzzy Logic Control

Hayatal Falah, Agus (2019) Implementasi Sensor Gas Dan Suara Pada Pengendalian Proses Sangrai Kopi Menggunakan Fuzzy Logic Control. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 07111750040010_Master_Thesis.pdf Restricted to Repository staff only Download (4MB)

Abstract

ualitas rasa kopi ditentukan oleh proses sangrai biji kopi. Kualitas rasa biji kopi yang sempurna dapat dicapai dengan suhu yang stabil selama proses sangrai. Dalam penelitian ini, sistem kontrol dirancang menggunakan metode Fuzzy Logic Control (FLC) untuk menjaga kestabilan proses sangrai biji kopi. Sistem kontrol suhu menggunakan tiga jenis setpoint, yaitu light, medium, dan dark, dengan suhu masing-masing sebesar 190 ℃, 230 ℃, dan 250 ℃ serta konsentrasi gas yang berbeda. Suhu dan gas yang dideteksi oleh sensor MQ-3 dipantau untuk menentukan tingkat kematangan biji kopi. Variabel lain yang dideteksi adalah suara retakan (cracking sound) menggunakan mikrofon EM-41. Suara retakan dikonversi ke domain frekuensi menggunakan Fast Fourier Transform (FFT) dan diidentifikasi menggunakan Neural Network (NN). Hasil percobaan menunjukkan bahwa pada setpoint light, proses sangrai berhenti secara otomatis pada menit ke-23:58 dengan dua kali retakan. Pada setpoint medium, proses sangrai berhenti secara otomatis pada menit ke-27:45 dengan 17 kali retakan. Sementara itu, pada setpoint dark, proses sangrai berhenti secara otomatis pada menit ke-28:00 dengan 28 kali retakan.
===================================================================================================================================
The flavor quality of coffee is largely determined by the roasting process. Optimal flavor can be achieved by maintaining a stable roasting temperature. This study developed a temperature control system based on the Fuzzy Logic Control (FLC) method to regulate the coffee bean roasting process. The system employs three temperature setpoints—light, medium, and dark—corresponding to 190 ℃, 230 ℃, and 250 ℃, respectively, each associated with different gas concentration levels. Temperature and gas concentration detected by an MQ-3 sensor are monitored to determine the roasting level of the coffee beans. Another important parameter is the cracking sound, which is captured using an EM-41 microphone. The recorded crack sounds are transformed into the frequency domain using the Fast Fourier Transform (FFT) and classified using a Neural Network (NN). Experimental results showed that at the light setpoint, the roasting process stopped automatically after 23 minutes and 58 seconds, with two crack events detected. At the medium setpoint, the process stopped after 27 minutes and 45 seconds, with 17 crack events detected. At the dark setpoint, roasting was completed automatically after 28 minutes, with 28 crack events detected.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	coffee roaster, fast fourier transform, fuzzy logic control, gas sensor, microphone
Subjects:	T Technology > TA Engineering (General). Civil engineering (General)
Divisions:	Faculty of Electrical Technology > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Mr Agus Falah
Date Deposited:	05 Aug 2026 08:01
Last Modified:	05 Aug 2026 08:01
URI:	http://repository.its.ac.id/id/eprint/67423

Actions (login required)

View Item