Sijabat, Fyio Frentina (2024) State Estimator Telemetering Untuk Superisory Control And Acquisition Data (SCADA) Transmisi Menggunakan Metode Artificial Neural Network. Diploma thesis, Institut Teknologi Sepuluh Nopember.
Text
Buku Tugas Akhir Fyio Frentina Sijabat_Departemen Teknik Elektro Otomasi.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (5MB) | Request a copy |
|
Text
2040201023_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (5MB) | Request a copy |
Abstract
Penelitian ini bertujuan untuk meningkatkan akurasi prediksi energi (kWh) dalam sistem transmisi Gardu Induk Patuha 150kV menggunakan metode Artificial Neural Network (ANN). Sistem tenaga listrik memerlukan sistem pengawasan, pengendalian, dan pengolahan data secara real time untuk meningkatkan kualitas pelayanan dan produk. SCADA, sebagai sistem kontrol proses, kerap mengalami gangguan pembacaan yang disebabkan oleh bertambahnya kapasitas gardu induk sehingga diperlukan state estimator yang lebih akurat. Metode ANN dipilih karena kemampuannya untuk memberikan hasil prediksi yang lebih optimal dibandingkan metode konvensional lainnya. Penelitian ini fokus pada prediksi energi dengan variabel input Arus (A), Daya Aktif (kW), dan Daya Reaktif (kVAr) yang dianalisis menggunakan ANN. Hasil pengujian menunjukkan bahwa variasi pembagian data, jumlah hidden layer, jumlah epoch, dan ukuran batch size berpengaruh signifikan terhadap akurasi model ANN. Nilai MAE terendah sebesar 0,18 dicapai dengan konfigurasi 128 neuron di hidden layer 1 dan 64 neuron di hidden layer 2. Dibandingkan dengan model linear regression dan decision tree regression, ANN menunjukkan akurasi prediksi yang lebih baik. Dengan demikian, penggunaan ANN dalam penelitian ini berhasil meningkatkan akurasi state estimator dan mengoptimalkan operasi sistem transmisi 150kV.
==================================================================================================================================
This research aims to improve the accuracy of energy prediction (kWh) in the Patuha 150kV Substation transmission system using the Artificial Neural Network (ANN) method. Electric power systems require supervision, control, and data processing systems in real time to improve the quality of services and products. SCADA, as a process control system, often experiences reading disturbances caused by increasing substation capacity so that a more accurate state estimator is needed. The ANN method was chosen because of its ability to provide more optimal prediction results than other conventional methods. This research focuses on energy prediction with input variables of Current (A), Active Power (kW), and Reactive Power (kVAr) analyzed using ANN. The test results show that variations in data division, number of hidden layers, number of epochs, and batch size have a significant effect on the accuracy of the ANN model. The lowest MAE value of 0.18 is achieved with a configuration of 128 neurons in hidden layer 1 and 64 neurons in hidden layer 2. Compared to linear regression and decision tree regression models, ANN shows better prediction accuracy. Thus, the use of ANN in this study successfully improved the accuracy of the state estimator and optimized the operation of the 150kV transmission system.
Item Type: | Thesis (Diploma) |
---|---|
Uncontrolled Keywords: | Artificial Neural Network, Supervisory Control And Data Acquisition, State Estimator, Transmisi, Transmission |
Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1001 Production of electric energy or power T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1007 Electric power systems control T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1010 Electric power system stability. Electric filters, Passive. |
Divisions: | Faculty of Vocational > 36304-Automation Electronic Engineering |
Depositing User: | FYIO FRENTINA SIJABAT |
Date Deposited: | 19 Aug 2024 01:50 |
Last Modified: | 19 Aug 2024 01:50 |
URI: | http://repository.its.ac.id/id/eprint/115459 |
Actions (login required)
View Item |