Identifikasi Beban Pada Sistem Tegangan Rendah Menggunakan Harmonisa Secar Real Time Berbasis Fast Fourier Transform dan Neural Network

Rasyiq, Naufal (2018) Identifikasi Beban Pada Sistem Tegangan Rendah Menggunakan Harmonisa Secar Real Time Berbasis Fast Fourier Transform dan Neural Network. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
07111340000135-Non_Degree.pdf - Published Version
Restricted to Repository staff only

Download (1MB) | Request a copy

Abstract

Semakin meningkatnya penggunaan beban non linear menimbulkan masalah pada sistem tenaga listrik. Beban non linear berpengaruh negatif terhadap sistem tenaga listrik seperti memperpendek usia peralatan dan mempercepat kerusakan peralatan listrik. Bagaimanapun, KWh meter yang terpasang pada pelanggan listrik PLN hanya menampilkan konsumsi energi listrik. Pada tugas akhir ini, dilakukan identifikasi jenis beban menggunakan metode backpropagation neural network. Proses pengumpulan data beban yang mengandung nilai harmonik dengan proses FFT (Fast Fourier Transform) menggunakan Mikrokontroller ARM STM32F7 yang dihubungkan dengan 5 jenis beban dipasang secara pararel. Backpropagation neural network akan dilatih untuk menyerap informasi untuk mengidentifikasi jenis beban/perangkat yang sedang digunakan dari nilai harmonik arusnya. Hasil dari tahap pelatihan ini akan menghasilkan source code yang akan dimasukkan kedalam mikro yang sudah terpasang di alat smart meter yang berguna di masa yang akan datang untuk keperluan pemantauan kualitas daya dan bisa mengetahui informasi daya total yang dihasilkan oleh alat rumah tangga tersebut. ============== The increasing use of non-linear loads creates problem with the power system. Non-linear loads negatively affect electrical affect electrical power system suchas shorthening the life of equipment and accelerating electrical equipment damage. Howefer, the KWh meter installed on PLN's electricity customers only displays electrical energy consumption. In this final project, identification of loads types using harmonic current values generated by household appliances using backpropagation neural network method. The process of collecting load data containing harmonic values with the FFT (Fast Fourier Transform) process using ARM STM32F7 microcontroller which is connected to the 5 types of loads installed parallel. Backpropagation neural network will be trained to absorb information to identify the type of load/devise being used from its harmonic current value. The result of this training phase will generate source code that will be incorporated into micro that has been installed in the smart meters tool that is useful in the future for the purpose of monitoring the quality of power and can know the total power information generated by the household appliance.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Fast Fourier Transform, arus harmonisa, neural network, backpropagation, Harmonic Current
Subjects: Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Q Science > QC Physics > QC20.7.F67 Fourier transformations
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3226 Transients (Electricity). Electric power systems. Harmonics (Electric waves).
Divisions: Faculty of Electrical Technology > Electrical Engineering > (S1) Undergraduate Theses
Depositing User: Naufal Rasyiq
Date Deposited: 09 Oct 2018 03:33
Last Modified: 09 Oct 2018 03:33
URI: http://repository.its.ac.id/id/eprint/52522

Actions (login required)

View Item View Item