Pengembangan sistem stabilisasi tegangan generator menggunakan ANN-PID dalam menghadapi perubahan beban listrik secara dinamis

Saputra, Muzaidin (2025) Pengembangan sistem stabilisasi tegangan generator menggunakan ANN-PID dalam menghadapi perubahan beban listrik secara dinamis. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 10311910000043-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Dalam sebuah pembangkit listrik, kestabilan tegangan menjadi faktor utama agar perangkat elektronik bisa beroperasi dengan lancar dan distribusi daya tetap efisien. Penelitian ini membahas penerapan metode Artificial Neural Network-Proportional Integral Derivative (ANN-PID) pada remote laboratory untuk mengendalikan tegangan keluaran generator dalam menghadapi perubahan beban dinamis. ANN digunakan untuk menyesuaikan parameter PID secara otomatis berdasarkan data latih yang diperoleh dari pengujian sistem PID konvensional. Metodologi yang digunakan meliputi perancangan perangkat keras, pengumpulan data sensor, perekaman data steady-state, serta implementasi dan pengujian ANN-PID. Dari pengujian sistem PID, diperoleh parameter optimal Kp = 5.87, Ki = 25.5, dan Kd = 5.55 dengan menggunakan fine tuning yang menunjukkan performa terbaik pada beban 3 lampu dengan tegangan rata-rata 5.02 V dan error 0.18 V. Selanjutnya, implementasi ANN-PID menunjukkan keunggulan dalam kestabilan. Pada variasi beban 1 dan 2 lampu, ANN-PID mampu mencapai kondisi steady-state dalam 1 detik, lebih cepat dibandingkan metode PID yang memerlukan waktu hingga 4 detik. Hasil ini menunjukkan bahwa metode ANN-PID mampu meningkatkan kestabilan tegangan sekaligus mempercepat respon sistem secara signifikan, menjadikannya solusi yang efektif untuk pengendalian generator pada remote laboratory dengan beban dinamis.
======================================================================================================================================
In a power plant, voltage stability is a key factor in ensuring that electronic devices operate smoothly and power distribution remains efficient. This study discusses the application of the Artificial Neural Network-Proportional Integral Derivative (ANN-PID) method in a remote laboratory to control generator output voltage in response to dynamic load changes. ANN is used to automatically adjust PID parameters based on training data obtained from testing the conventional PID system. The methodology employed includes hardware design, sensor data collection, steady-state data recording, as well as the implementation and testing of the ANN-PID system. From the PID system testing, optimal parameters were obtained: Kp = 5.87, Ki = 25.5, and Kd = 5.55 using fine-tuning, showing the best performance at a 3-light load with an average voltage of 5.02 V and an error of 0.18 V. Furthermore, the implementation of ANN-PID demonstrated superior stability. At load variations of 1 and 2 lamps, the ANN-PID was able to achieve steady-state conditions in 1 second, faster than the PID method, which required up to 4 seconds. These results indicate that the ANN-PID method can improve voltage stability while significantly accelerating system response, making it an effective solution for generator control in remote laboratories with dynamic loads.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	ANN-PID, kestabilan tegangan, respon sistem, remote laboratory ANN-PID, voltage stability, system response, remote laboratory
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1001 Production of electric energy or power 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:	Muzaidin Saputra
Date Deposited:	07 Aug 2025 06:10
Last Modified:	07 Aug 2025 06:10
URI:	http://repository.its.ac.id/id/eprint/127928

Actions (login required)

View Item