Peramalan Beban Listrik Jangka Pendek Menggunakan Optimally Pruned Extreme Learning Machine (OPELM) Pada Sistem Kelistrikan Jawa Timur - Short-Term Load Forecasting Using Optimally Pruned Extreme Learning Machine (OPELM) in Power System at East Java

Perdana, Januar Adi (2012) Peramalan Beban Listrik Jangka Pendek Menggunakan Optimally Pruned Extreme Learning Machine (OPELM) Pada Sistem Kelistrikan Jawa Timur - Short-Term Load Forecasting Using Optimally Pruned Extreme Learning Machine (OPELM) in Power System at East Java. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
2208100150-Undergraduate Thesis.pdf - Published Version

Download (1MB) | Preview

Abstract

Peramalan beban listrik jangka pendek merupakan faktor yang sangat penting dalam perencanaan dan pengoperasian sistem tenaga listrik. Tujuan dari peramalan beban listrik adalah agar permintaan listrik dan penyediaan listrik dapat seimbang. Karakteristik beban di wilayah Jawa Timur sangat fluktuatif sehingga pada penelitian ini digunakan metode Optimally pruned extreme learning machine (OPELM) untuk meramalkan beban listrik. Kelebihan OPELM ada pada learning speed yang cepat dan pemilihan model yang tepat meskipun datanya mempunyai pola non linier. Keakuratan metode OPELM dapat diketahui dengan menggunakan metode pembanding yaitu metode ELM. Kriteria keakuratan yang digunakan adalah MAPE. Hasil dari perbandingan kriteria keakuratan menunjukkan bahwa hasil peramalan OPELM lebih baik dari ELM. Error rata-rata hasil pengujian peramalan paling minimum menunjukkan MAPE sebesar 1,5534% terjadi pada peramalan hari Jumat, sementara pada hari yang sama dengan metode ELM menghasilkan MAPE sebesar 2,6832%. ======================================================================================================================== Short-term electric load forecasting is a very important factor for planning and operation of electric power systems. The purpose of the electrical load forecasting is that electricity demand and supply can be balanced. Load characteristics in East Java is very fluctuative, so in this research used methods of optimally pruned extreme learning machine (OPELM) to forecast electricity load. Advantadges OPELM on a fast learning speed and the selection of an appropriate model although the data have non-linear pattern. OPELM accuracy can be determined by using the comparison ELM method. Accuracy criteria used MAPE. The results of the comparison criteria indicate that the forecasting accuracy of OPELM is better than ELM. Result of testing on forecast gives the smallest average error percentage by MAPE of 1,5534% for forecasting Friday, while on the same day forecasting, ELM give MAPE of 2,6832%.

Item Type: Thesis (Undergraduate)
Additional Information: RSE 621.374 5 Per p
Uncontrolled Keywords: Peramalan Beban Listrik Jangka Pendek, OPELM, ELM, Short-term electric load forecasting, OPELM, ELM.
Subjects: T Technology > T Technology (General) > T174 Technological forecasting
Divisions: Faculty of Industrial Technology > Electrical Engineering
Depositing User: ansi aflacha
Date Deposited: 12 Dec 2019 07:09
Last Modified: 12 Dec 2019 07:09
URI: http://repository.its.ac.id/id/eprint/72346

Actions (login required)

View Item View Item