Rancang Bangun Sistem Switching Daya Otomatis Pada Beban Kontinu Berbasis Algoritma Decision Tree

Syifa, Alifian Asy (2026) Rancang Bangun Sistem Switching Daya Otomatis Pada Beban Kontinu Berbasis Algoritma Decision Tree. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2042221137_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Beban listrik kontinu berperan penting di berbagai sektor, seperti aerator budidaya ikan, pompa nutrisi hidroponik, dan sistem pendingin makanan. Namun, ketergantungan pada suplai listrik PLN menyebabkan sistem rentan terganggu saat terjadi pemadaman. Untuk mengatasi permasalahan tersebut dalam penelitian ini telah dirancang sistem switching daya otomatis berbasis algoritma Decision Tree untuk mengelola peralihan sumber energi antara Pembangkit Listrik Tenaga Surya (PLTS) dan listrik PLN dengan mempertimbangkan parameter State of Charge (SOC) baterai, tegangan PLN, tegangan inverter, dan arus beban. Sistem dikendalikan oleh mikrokontroler ESP32 yang dilengkapi sensor INA219 dan PZEM-004T serta Automatic Transfer Switch (ATS) berbasis Solid State Relay, dengan pemantauan data secara real-time melalui antarmuka berbasis Android. Model Decision Tree dikembangkan menggunakan library scikit-learn dalam bahasa pemrograman Python, kemudian dikonversi dan di-embedded secara lokal pada mikrokontroler ESP32-S3 untuk dijalankan secara langsung. Hasil pengujian menunjukkan bahwa sensor INA219 memiliki akurasi 97,52%, sedangkan sensor PZEM-004T pada sisi PLN dan inverter masing-masing mampu mengukur tegangan dengan akurasi 99,80% dan 99,75%, serta arus sebesar 95,60% dan 95,65%. Model Decision Tree berhasil menghasilkan tingkat akurasi 85,00%, di mana respons logika aktual yang telah di-embedded pada ESP32-S3 sepenuhnya selaras dengan output ideal model Decision Tree. Sementara itu, sistem switching daya otomatis menunjukkan proses perpindahan sumber energi yang stabil dan konsisten dengan waktu dead time rata-rata 5,035 detik dari PLN ke PLTS dan 5,002 detik dari PLTS ke PLN.
=========================================================================================================================================
Continuous electrical loads play a crucial role across various sectors, such as aquaculture aerators, hydroponic nutrient pumps, and food refrigeration systems. However, reliance on the main grid (PLN) leaves these systems vulnerable to power outages. This research has successfully designed and tested an automatic power switching system based on the Decision Tree algorithm to manage energy source transitions between a Solar Power System (PLTS) and the main grid, considering parameters such as battery State of Charge (SOC), grid voltage, inverter voltage, and load current. The system is controlled by an ESP32 microcontroller equipped with INA219 and PZEM-004T sensors, as well as a Solid State Relay-based Automatic Transfer Switch (ATS), with real-time data monitoring via an Android-based interface. The Decision Tree model was developed using the scikit-learn library in Python, then converted and locally embedded into the ESP32-S3 microcontroller for direct execution. Test results show that the INA219 sensor achieved an accuracy of 97.52%, while the PZEM-004T sensors on the grid and inverter sides measured voltage with accuracies of 99.80% and 99.75%, and current with accuracies of 95.60% and 95.65%, respectively. The Decision Tree model achieved an accuracy rate of 85.00%, with the actual embedded logic response on the ESP32- S3 fully aligning with the ideal model output. Meanwhile, the automatic power switching system demonstrated a stable and consistent transition process with an average dead time of 5.035 seconds from grid to PLTS and 5.002 seconds from PLTS to grid.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Automatic Transfer Switch, Decsion Tree, IoT, PLTS, Automatic Transfer Switch, Decision Tree, IoT, Solar energy
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1007 Electric power systems control T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5103.8 Switching systems
Divisions:	Faculty of Vocational > Instrumentation Engineering
Depositing User:	Alifian Asy Syifa
Date Deposited:	15 Sep 2026 04:12
Last Modified:	15 Sep 2026 04:12
URI:	http://repository.its.ac.id/id/eprint/144496

Actions (login required)

View Item