ANALISIS GANGGUAN PADA JARINGAN DISTRIBUSI BERBASIS FUZZY LOGIC DAN JARINGAN SYARAF TIRUAN

• Main Author: Devi Ivana Athaliana, - (Author)
• Format: Book
• Published: 2020-07-23.
• Subjects: Thesis; NonPeerReviewed
• Online Access: Link Metadata
• Notes: http://repository.upi.edu/50021/8/S_TE_1607331_Title.pdf; http://repository.upi.edu/50021/2/S_TE_1607331_Chapter1.pdf; http://repository.upi.edu/50021/3/S_TE_1607331_Chapter2.pdf; http://repository.upi.edu/50021/4/S_TE_1607331_Chapter3.pdf; http://repository.upi.edu/50021/5/S_TE_1607331_Chapter4.pdf; http://repository.upi.edu/50021/6/S_TE_1607331_Chapter5.pdf; http://repository.upi.edu/50021/7/S_TE_1607331_Appendix.pdf
• Keywords: L Education (General); TK Electrical engineering. Electronics Nuclear engineering

Jaringan distribusi merupakan jaringan yang paling dekat dengan konsumen dan tercatat memiliki banyak jumlah gangguan. Pendeteksian gangguan dengan akurat dapat menimalisir dampak gangguan untuk penyaluran listrik yang berkelanjutan. Pada penelitian ini pendeteksian jenis gangguan dilakukan menggunakan fuzzy logic dan pendeteksian lokasi gangguan dilakukan menggunakan jaringan syaraf tiruan. Metoda berikut dipilih karena memiliki keunggulan dalam kecepatan dan ketepatan dalam pendeteksian dibandingakan dengan metode konvensional. Berdasarkan hasil penelitian, fuzzy logic dapat mendeteksi jenis gangguan dengan baik dan jaringan syaraf tiruan memiliki akurasi sebesar 88,7% untuk mendeteksi lokasi gangguan. Distribution Network is the network that closest to consumers and has a large number of fault. By detecting faults accurately, can minimalize the impact of continuous electricity distribution. In this research, the type of fault detection perform using fuzzy logic and location detection perform using artificial neural network. The following method is chosen because it can detect both quickly and precisely compared with the conventional method. Based on the result of the research, fuzzy logic can perform well for detecting fault type, besides artificial neural networks have an 88,7% accuracy for detecting fault location.