Perancangan Penerapan Optimasi Algoritma pada Sistem Interkoneksi Jaringan Distribusi Energi Listrik

Kafrawi M. Tuara, Ahmad Agus Setiawan, ST., M.Sc., Ph.D.; Prof (Emr). Adhi Susanto, M.Sc., Ph.D.

2018 | Tesis | S2 TEKNIK SISTEM

Penelitian ini menggunakan pendekatan analisis dengan dua algoritma optimasi untuk merancang perbandingan optimasi topologi interkoneksi pada sistem jaringan distribusi tenaga listrik. Sesuai eksplorasi dan analisis terhadap jarak optimal terhadap setiap sentral wilayah operasional PT. PLN (Persero) Tbk. Maka, dipergunakanlah suatu analisis Logaritma Alami (natural logarithms), serta analisis Matriks Diagram Nol (extract triangular matrix) untuk menghasilkan topologi interkoneksi optimal terhadap setiap lintasan sistem jaringan distribusi tenaga listrik. Hasil prosentase karakteristik menunjukan bahwa proses optimasi perbandingan adalah 100% oleh Algortima Genetika dan prosentase 98% oleh MoTSA's. Untuk pengembangan analisis dan menilai hasil kinerja dari metode optimasi Algoritma Genetika, pengujian (testable) terhadap karakterisasi data-data yang dioptimasi kemudian dikompilasi kedalam bentuk Matrik Diagram nol. Pada akhirnya jarak interkoneksi minimum diperoleh secara keseluruhan adalah 29,93 Kmc. Secara total hasil optimasi menemukan 6 wilayah sentral operasional dengan jalur/rute terhubung pada sistem interkoneksi jaringan distribusi tenaga listrik adalah sebanyak 25 Kota sebagai Sub-Sentral transit distribusi tenaga listrik. Hasil perbandingan optimasi diperoleh jarak interkoneksi sistem jaringan distribusi tenaga listrik sepanjang 5.096,33 Kmr menggunakan Algoritma Genetika (AG), sedangkan jarak 119,43 Kmr menggunakan Algoritma Tiga Langkah (Method of the Three Steps Algorithm, MoTSA's). Selanjutnya perbedaan kedua algoritma yang digunakan dalam penelitian ini adalah pada proses optimasi. Dimana, metode AG perlu diberikan beberapa kali iterasi serta melalui beberapa kali tahap analisis untuk menghasilkan nilai optimum. Sedangkan optimasi menggunakan Metode MoTSA's hanya dilakukan (running program) dalam satu kali proses optimasi saja. Oleh karena itu, sasaran capaian 35.000 MW ditingkatkan menjadi 1,017 TWh pada tahun 2034 dalam proyeksi Program Pembangunan Nasional oleh Pemerintah dapat terealisasi secara maksimal.

The existence of the development of power generation policy to send forth the electrical energy from the region have a highly primary energy sources to the others areas where have limited resources. The transmission system as supporting framework of the electrical distribution system should be requiring a strong interconnection networks (backbone) are given the wide geographical distance among the regions. In this study, the analytical approaches by the two of algorithms optimization are used to engineering design comparatively to the forms of topology configurations to the electrical distribution networks. Totally, the comparison yields of the algorithms optimization to the networks interconnection optimum through all to the central's area connectivity's in Indonesia has achieved the optima distances about 5096.33 Kmc of Genetic Algorithms (GA's), and by the comparative algorithm of MoTSA's is 119.43 Kmc about. Whereas, the characteristic percentages of the optimization outputs by these comparatively are 100% of GA's and 98% of MoTSA's. This study analyses the potential of the implementation Genetic Algorithms (GA's) to purpose a new strategy of the networks optimization which was originated from the idea of the routes length interconnection problem especially to the electricity distribution networks system in Indonesia. GA's have received a considerable attention in evolutionary computation. Thereby to develop and assess the performance outputs of GA's optimization by testable and the closed form characterization of Extract Triangular Matrix (ETM) analysis, which was found the six main centrals interconnection routes to 25 cities at wider areas of PLN's operational with the overall minimum distance is 29.93 Kmc. Finally, the existence of this study is providing an overview of the electrical distribution networks system based on the authentically data's by the PLN's, it is expected to become a reference material for the completion of the availability and utilization of the electrical energy in terms of installation and the network withdrawal aspects. The results of this study are also expected to be used as an input data to the electrical demand policy according to the concepts of the PLN's RUKN draft 2015-2034. Thereby, the goal's target of 35.000 MW for National Development Program projection to 1.017 TWh in 2034 of the electricity demand would be keeping down.

Kata Kunci : Algoritma Genetika, MoTSA's, Matriks Diagram Nol, Sistem Distribusi Listrik, Topologi Interkoneksi, Teknik Optimasi