Tefnai, Abdu Rachman Algafikki (2026) Peningkatan Kinerja Sistem Distribusi Tenaga Listrik melalui Prediksi Susut dan Evaluasi Konfigurasi Jaringan Menggunakan Artificial Neural Network. Masters thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 6022241115-Master_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
6022241115-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (14MB) | Request a copy |
Abstract
Susut distribusi merupakan indikator utama efisiensi sistem distribusi tenaga listrik karena berdampak langsung pada kehilangan energi, kualitas tegangan, dan kinerja keuangan penyedia layanan. Penelitian ini mengembangkan model prediksi susut distribusi berbasis Jaringan Saraf Tiruan (Artificial Neural Network/ANN) sebagai dasar evaluasi rekonfigurasi jaringan pada sistem distribusi PT PLN (Persero) ULP Boja. Data penelitian bersumber dari data operasional PLN ULP Boja periode Januari 2023 hingga Desember 2024. Variabel input model terdiri dari panjang penyulang, beban siang, beban malam, tegangan, dan beban rata-rata, sedangkan variabel output berupa persentase susut distribusi. Data melalui tahap pembersihan, penanganan nilai hilang, analisis pencilan, dan normalisasi Min-Max sebelum dibagi menjadi 70% data pelatihan, 15% data validasi, dan 15% data pengujian. Model terbaik menghasilkan arsitektur input-15-10-1 dengan fungsi aktivasi ReLU pada lapisan tersembunyi dan lapisan output linier. Hasil pengujian menunjukkan nilai MSE sebesar 1,145763; RMSE 1,070403; MAE ±0,892000; MAPE 14,870427%; dan $R^{2}$ 0,048314. Nilai tersebut menunjukkan ANN layak digunakan sebagai alat estimasi awal susut daya, meskipun keputusan operasional tetap memerlukan verifikasi melalui simulasi aliran daya. Hasil evaluasi rekonfigurasi jaringan menunjukkan penurunan susut aktif menjadi 889,3 kW, susut reaktif menjadi 6.251,9 kvar, peningkatan tegangan minimum dari 72,5% menjadi 92,2%, serta penurunan jatuh tegangan maksimum dari 23,47% menjadi 3,98%.
=====================================================================================================================================
Distribution loss is a key indicator of power distribution system efficiency because it directly affects energy losses, voltage quality, and utility financial performance. This research develops an Artificial Neural Network (ANN)-based distribution loss prediction model as a basis for evaluating network reconfiguration in the distribution system of PT PLN (Persero) ULP Boja. The dataset was obtained from operational data of PLN ULP Boja from January 2023 to December 2024, with input variables consisting of feeder length, daytime load, nighttime load, voltage, and average load, while the model output was the distribution loss percentage. The data were processed through cleaning, missing value treatment, outlier analysis, and Min-Max normalization before being divided into 70% training, 15% validation, and 15% testing subsets. The best model used an input-15-10-1 architecture with ReLU activation in the hidden layers and a linear output layer. Testing results yielded an MSE of 1.145763, RMSE of 1.070403, MAE of approximately 0.892000, MAPE of 14.870427%, and R2 of 0.048314. These values indicate that ANN is suitable as an initial loss estimation tool, while operational decisions still require verification through power-flow simulation. Network reconfiguration evaluation showed that active losses decreased to 889.3 kW, reactive losses decreased to 6,251.9 kvar, the minimum voltage increased from 72.5% to 92.2%, and the maximum voltage drop decreased from 23.47% to 3.98%.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Susut Distribusi, Jaringan Syaraf Tiruan, Rekonfigurasi Jaringan, Sistem Distribusi Tenaga Listrik, Simulasi Aliran Daya, Efisiensi Energi, ULP Boja, Artificial Neural Network, distribution loss, network reconfiguration, distribution system |
| Subjects: | Q Science > Q Science (General) > Q325.78 Back propagation Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA402 System analysis. |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis |
| Depositing User: | Abdu Rachman Algafikki Tefnai |
| Date Deposited: | 31 Jul 2026 03:25 |
| Last Modified: | 31 Jul 2026 03:25 |
| URI: | http://repository.its.ac.id/id/eprint/140383 |
Actions (login required)
 |
View Item |