Analisis Sistem Distribusi Listrik Wilayah PT. PLN (Persero) ULP Surakarta Kota Menggunakan Metode Adaptive Modified Firefly Algorithm Untuk Optimasi Manuver Beban

Hamiseno, Prasetio (2026) Analisis Sistem Distribusi Listrik Wilayah PT. PLN (Persero) ULP Surakarta Kota Menggunakan Metode Adaptive Modified Firefly Algorithm Untuk Optimasi Manuver Beban. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6022241054-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (17MB)

Abstract

Gangguan pada pemutus tenaga feeder dapat menghentikan suplai ke sebagian pelanggan meskipun jaringan distribusi loop masih memiliki jalur alternatif. Pemilihan status peralatan hubung yang kurang tepat dapat memindahkan beban secara berlebihan, menambah rugi-rugi daya, atau menyebabkan saluran melewati batas pembebanan. Penelitian ini mengembangkan optimasi transfer beban pada sistem distribusi loop 20 kV menggunakan Adaptive Modified Firefly Algorithm (AMFA) yang terintegrasi dengan Python dan DIgSILENT PowerFactory. Enam load break switch dimodelkan sebagai variabel keputusan biner, sedangkan PMT feeder yang mengalami gangguan dipertahankan terbuka dan PMT feeder sehat dikunci tertutup. Fungsi objektif disusun secara bertingkat dengan mendahulukan konfigurasi yang valid dan layak, meminimalkan jumlah general load yang tidak tersuplai, kemudian meminimalkan rugi-rugi daya aktif. Kandidat ditolak apabila aliran daya tidak konvergen atau maximum line loading mencapai 90% atau lebih. Empat skenario gangguan diterapkan pada PMT Feeder MKN 1, MKN 4, MKN 5, dan MKN 10. Kinerja AMFA dibandingkan dengan Genetic Algorithm (GA), Binary Particle Swarm Optimization (BPSO), dan Firefly Algorithm (FA) menggunakan model serta batasan kelistrikan yang sama. Hasil simulasi menunjukkan bahwa seluruh metode berhasil memulihkan 14 dari 14 general load pada setiap skenario. Rugi-rugi daya akhir 0,143410 MW dengan kapasitas line loading tidak lebih dari 90%. Keempat menghasilkan kualitas solusi akhir yang sama, tetapi memiliki proses pencarian dan waktu komputasi yang berbeda. AMFA mencatat runtime dengan rata-rata 0,233 detik, serta menemukan solusi akhir pada rata-rata evaluation sequence ke-18. Konfigurasi hasil optimasi metode AMFA dapat digunakan sebagai pendukung keputusan manuver beban yang paling tepat dan cepat.
=========================================================================================================================================
A fault at a feeder circuit breaker can interrupt the electricity supply to some customers even when an alternative supply path remains available in a loop distribution network. An inappropriate switching configuration may cause excessive load transfer, increase power losses, or push line loading beyond its allowable limit. This study develops a load-transfer optimization method for a 20 kV loop distribution system using the Adaptive Modified Firefly Algorithm (AMFA) integrated with Python and DIgSILENT PowerFactory. Six load break switches are represented as binary decision variables, while the circuit breaker of the faulted feeder is maintained in the open position and the circuit breakers of healthy feeders are locked in the closed position. A hierarchical objective function is formulated to prioritize valid and feasible configurations, minimize the number of unsupplied general loads, and subsequently minimize active power losses. A candidate configuration is rejected when the load-flow calculation fails to converge or when the maximum line loading reaches or exceeds 90%. Four fault scenarios are examined at the circuit breakers of Feeders MKN 1, MKN 4, MKN 5, and MKN 10. The performance of AMFA is compared with the Genetic Algorithm (GA), Binary Particle Swarm Optimization (BPSO), and Firefly Algorithm (FA) using the same network model and electrical constraints. The simulation results show that all four methods successfully restored all 14 general loads under every fault scenario. The lowest final active power loss was 0.143410 MW, while the maximum line loading remained below 90% in all cases. Although the four methods produced the same final solution quality, they differed in their search processes and computational times. AMFA achieved an average runtime of 0.233 seconds and identified the final solution at an average evaluation sequence of 18. These findings indicate that the switching configuration obtained using AMFA can serve as a reliable and timely decision-support reference for load-transfer operations.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Adaptive Modified Firefly Algorithm, rekonfigurasi jaringan distribusi, transfer beban, restorasi pelayanan, rugi-rugi daya, Adaptive Modified Firefly Algorithm, distribution network reconfiguration, load transfer, service restoration, power loss.
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3030 Electric power distribution systems
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Prasetio Hamiseno
Date Deposited:	01 Aug 2026 02:11
Last Modified:	01 Aug 2026 02:11
URI:	http://repository.its.ac.id/id/eprint/142783

Actions (login required)

View Item