Imran, Andi Imran (2021) Prediksi Optimal Beban Puncak Hari Libur Nasional Berbasis Interval Type 2 Fuzzy Inference System - Firefly (Studi Kasus Jawa-Bali). Doctoral thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 02 Buku Disertasi Andi Imran.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
02 Buku Disertasi Andi Imran.pdf - Accepted Version Restricted to Repository staff only Download (1MB) | Request a copy |
Abstract
ABSTRAK
Peramalan merupakan taksiran ilmiah yang bertujuan untuk memperkirakan apa yang akan terjadi pada masa yang akan datang. Penelitian ini membahas peramalan beban jangka pendek yaitu beban puncak hari libur nasional. Peramalan dilakukan dengan menggunakan metode Fuzzy Logic System Type 2 yaitu Interval Type-2 Fuzzy Inference System yang dioptimasi dengan firefly algorithm. Metode tersebut merupakan metode baru yang digunakan dalam peramalan beban puncak hari libur nasional.
Firefly algorithm digunakan untuk mengoptimalkan Footprint of Uncertainty (FOU) pada fuzzy logic yang terdiri dari antecedent (X, Y) dan consequent (Z). Salah satu keuntungan dari firefly algorithm mempunyai kinerja yang sangat baik dalam hal optimasi. Metode yang diusulkan menggunakan data dari beban puncak hari libur nasional pada sistem kelistrikan Jawa-Bali. Penelitian ini difokuskan pada data beban puncak dari empat hari sebelum hari libur (h-4) dan pada hari libur (h). Metode Interval Type 1 Fuzzy Inference System (IT1FIS), Interval Type 2 Fuzzy Inference System (IT2FIS), Interval Type 2 Fuzzy Inference System - Flower Pollination Algorithm (IT2FIS-FPA), dan Interval Type 2 Fuzzy Inference System - Cuckoo Search Algorithm (IT2FISCSA) dijadikan sbagai pembanding dalam penelitian ini.
Hasil pengujian menunjukkan bahwa metode Interval Type 2 Fuzzy Inference System - Firefly Algorithm (IT2FIS-FA) memberikan peramalan yang akurat dengan mean absolute percentage error (MAPE) dibawah 2%. MAPE peramalan beban puncak hari libur nasional menggunakan IT2FIS-FA sebesar 1,39836%, menggunakan IT1FIS sebesar 1,627703%, mengguanakan IT2FIS 1,69757%, menggunakan IT2FIS-FPA sebesar 1,52907% dan menggunakan IT2FISCSA sebesar 1,473775%.
Kata Kunci : Peramalan Beban Jangka Pendek, Interval Type-2 Fuzzy Inference System, Firefly Algorithm, Mean Absolute Percentage Error (MAPE)
==============================================================================================
Forecasting is a scientific estimate that aims to predict what will happen in the future. This study discusses short-term load forecasting, namely the peak load of national holidays. Forecasting is carried out using the Fuzzy Logic System Type 2 method, namely Interval Type-2 Fuzzy Inference System which is optimized with a firefly algorithm. The method is a new method used in forecasting the peak load of national holidays
The Firefly algorithm is used to optimize the Footprint of Uncertainty (FOU) on fuzzy logic which consists of antecedent (X, Y) and consequent (Z). One of the advantages of firefly algorithm is that it performs very well in terms of optimization. The proposed method uses data from the peak load of national holidays in the Java-Bali electricity system. This study focused on peak load data from four days before holidays (d-4) and on holidays (h). Interval Type 1 Fuzzy Inference System (IT1FIS), Interval Type 2 Fuzzy Inference System (IT2FIS), Interval Type 2 Fuzzy Inference System - Flower Pollination Algorithm (IT2FIS-FPA), and Interval Type 2 Fuzzy Inference System - Cuckoo Search Algorithm (IT2FISCSA) is used as a comparison in this study
The results show that the Interval Type 2 Fuzzy Inference System - Firefly Algorithm (IT2FIS-FA) method provides accurate forecasting with a mean absolute percentage error (MAPE) below 2%. The peak load national holidays forecasting MAPE using IT2FIS-FA amounted to 1.39836%, using IT1FIS of 1.627703%, using IT2FIS of 1,69757%, using IT2FIS-FPA of 1,52907% and using IT2FISCSA of 1,473775%.
Keyword : Short-Term Load Forecasting, Interval Type-2 Fuzzy Inference System, Firefly Algorithm, Mean Absolute Percentage Error (MAPE)
| Item Type: | Thesis (Doctoral) |
|---|---|
| Uncontrolled Keywords: | Kata Kunci : Peramalan Beban Jangka Pendek, Interval Type-2 Fuzzy Inference System, Firefly Algorithm, Mean Absolute Percentage Error (MAPE) Keyword : Short-Term Load Forecasting, Interval Type-2 Fuzzy Inference System, Firefly Algorithm, Mean Absolute Percentage Error (MAPE) |
| Subjects: | T Technology > T Technology (General) > T57.62 Simulation T Technology > T Technology (General) > T57.84 Heuristic algorithms. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1007 Electric power systems control |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20001-(S3) PhD Thesis |
| Depositing User: | Andi Imran |
| Date Deposited: | 05 Mar 2021 00:41 |
| Last Modified: | 05 Mar 2021 00:41 |
| URI: | http://repository.its.ac.id/id/eprint/83495 |
Actions (login required)
 |
View Item |