Rudiarsono, Prasetia (2018) PENERAPAN METODE JARINGAN SYARAF TIRUAN PROPAGASI BALIK UNTUK PERAMALAN BEBAN LISTRIK JANGKA PENDEK DI SISTEM KELISTRIKAN KOTA MALANG. Skripsi thesis, Institut Teknologi Nasional Malang.
Text
15. FILE LENGKAP SKRIPSI.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB) | Request a copy |
Abstract
Penerapan Metode Jaringan Syaraf Tiruan Propagasi Balik Untuk Peramalan Beban Listrik Jangka Pendek Di Sistem Kelistrikan Kota Malang Prasetia Rudiarsono I Made Wartana prasetiarudiarsono@gmail.com Abstrak Meningkatnya permintaan akan energi listrik menuntut produsen dapat memenuhi kebutuhan konsumen yang terus meningkat. Untuk mengatasi masalah tersebut, maka diperlukan peramalan beban listrik yang akurat. Penelitian ini akan membahas metode peramalan beban jangka pendek (short term load forecasting), yang merupakan peramalan beban dalam jangka waktu beberapa jam sampai satu minggu. Metode yang digunakan adalah jaringan syaraf tiruan (JST) backpropagation, karena kemampuan pendekatan yang baik terhadap ketidakliniearan. Untuk menguji efektivitasnya, maka dilakukan dua studi kasus, yaitu Studi kasus I menggunakan data input training beban dan suhu histori selama 2 minggu, sedangkan studi kasus II menggunakan data input training beban dan suhu histori selama 3 minggu. Data target dari dua studi kasus dan parameter-parameter training yang digunakan tetap sama. Dari hasil peramalan beban listrik tanggal 24 sampai 30 April 2018 di Gardu Induk Blimbing dan Kebonagung wilayah kota Malang untuk dua studi kasus tersebut, yaitu hasil peramalan beban listrik dari dua studi kasus mendekati beban aktual dengan tingkat akurasi sangat baik error dibawah 10% dengan persentase rata-rata error studi kasus I 7,24314%, tetapi studi kasus II memiliki error lebih rendah sebesar 5,480062%. Kata kunci – Peramalan beban listrik Jangka Pendek; Jaringan Syaraf Tiruan; Backpropagation 5 Implementation Of Artificial Neural Network Backpropagation For Short Term Load Forecasting In Malang City Prasetia Rudiarsono I Made Wartana prasetiarudiarsono@gmail.com Abstract The increasing of electricity demand is requires producers to meet the needs consumers. To resolve this problem, the accurate load forcasting is needed. This study will discuss the method of short term load forecasting, which is forecasting the load within a period of several hours to one week. The method that use is backpropagation Artificial Neural Network (ANN), because it has the ability of a good approach to nonlinearity. To test its effectiveness, two case were carried out, namely Case I that used input training data of load and temperature for 2 weeks, while case II using input training load data and temperature for 3 weeks. The target data and training parameters that used in two case were same. From the results of forecasting the electricity load on 24 to 30 April 2018 at Blimbing and Kebonagung substations in Malang city for the two case studies, the results of electricity load forecasting from two case studies were close to the actual load with a very good accuracy and error value is under 10% with an average percentage error of case I was 7,24314%, but in case II had a lower error, that was 5,480062%. Keywords: Short-Term Load Forecasting; Artificial Neural Network; Backpropagation
Item Type: | Thesis (Skripsi) |
---|---|
Uncontrolled Keywords: | Short-Term Load Forecasting; Artificial Neural Network; Backpropagation |
Subjects: | Engineering > Electrical Engineering |
Divisions: | Fakultas Teknologi Industri > Teknik Elektro S1 > Teknik Elektro S1(Skripsi) |
Depositing User: | Ms Nunuk Yuli |
Date Deposited: | 21 Feb 2019 02:59 |
Last Modified: | 13 Mar 2019 06:24 |
URI: | http://eprints.itn.ac.id/id/eprint/2226 |
Actions (login required)
View Item |