Prabowo, Rizky and Nurfadlilah, Zuliana and Lumbanraja, Favorisen R and Kurniawan, Didik (2021) DETEKSI TINGKAT KERUSAKAN SISTEM KELISTRIKAN PADA MOBIL MENGGUNAKAN ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM. Kumpulan Jurnal Ilmu Komputer (KLIK), 8 (1). ISSN 2406-7857
|
Text
377-862-1-PB.pdf Download (873kB) | Preview |
Abstract
The automotive industry in Indonesia has significant increase in the past decade. A famous car company opened a manufacturing branch to increase its production capacity in Indonesia. An increase in sales is directly proportional to an increase in service to customers. Damage on electrical system is the majority of modern car. Unfortunately, car users have minimal knowledge of car electricity. This article describes the technique of detecting the level of damage to a car's electrical system using the Adaptive Neuro-Fuzzy Inference System (Anfis) concept. As a case study in designing the system in question is the electrical system on the Toyota Avanza. Formation of a fuzzy inference system which is used for the system formation process through a GUI-based interface design (Graphic User Interface). The output of the system is a fuzzy analysis based on the membership function of the Gaussian, Triangular and Trapezoid methods to obtain an analysis of the level of damage to the electrical system on a Toyota Avanza. From the results of the system test for starter system, firewire system and lighting system, it is concluded that the analysis of the level of damage to the electrical system on the car using Anfis based on the Gaussian membership function model is more accurate(reach 85%) in predicting the level of damage to the analyzed electrical system.
| Item Type: | Article |
|---|---|
| Subjects: | Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Matematika dan Ilmu Pengetahuan Alam (FMIPA) > Prodi Ilmu Komputer |
| Depositing User: | rizky prabowo |
| Date Deposited: | 27 May 2021 02:34 |
| Last Modified: | 27 May 2021 02:34 |
| URI: | http://repository.lppm.unila.ac.id/id/eprint/30799 |
Actions (login required)
 |
View Item |