Pratama, Fysna Candra (2024) Pengembangan Energy Management System Pada Smart Building Terintegrasi Dengan Photovoltaic dan BESS Mempertimbangkan Prediksi Iradiasi Matahari Dan Prioritas Beban. Other thesis, Institut Teknologi Sepuluh Nopember.
![]() |
Text
5022211101-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (9MB) | Request a copy |
Abstract
Energi listrik merupakan elemen vital dalam operasional bangunan, dengan konsumsi mencapai 30-45% energi global. Smart building memanfaatkan energi terbarukan seperti panel surya, namun kinerjanya sangat bergantung pada cuaca, sehingga membutuhkan sistem penyimpanan energi (BESS) untuk menjaga stabilitas pasokan daya. Penelitian ini mengembangkan sistem manajemen energi berbasis algoritma Mixed Integer Linear Programming (MILP) dan memanfaatkan XGBoost untuk prediksi kebutuhan energi (load forecasting) dan forecast daya keluaran panel surya. Hasil prediksi yang akurat memungkinkan optimalisasi pemanfaatan panel surya dan BESS, sekaligus mengurangi ketergantungan pada jaringan listrik. Sistem diuji pada Gedung Chamchuri 5, Universitas Chulalongkorn, Thailand, dan Departemen Teknik Elektro ITS Surabaya, Indonesia. Pada Chamchuri 5, error prediksi keluaran daya panel surya mencapai MSE 302.47 Watt dan beban 2633.31 Watt, sedangkan di Teknik Elektro, error prediksi keluaran daya panel surya mencapai MSE 234.37 Watt dan beban 1721 Watt. Selain itu, Wireless Power Meter yang digunakan memiliki error tegangan sebesar 10.53 Volt dan arus 0.03 A. Smart switch sebagai perangkat keras yang terkontrol juga terintegrasi pada sistem manajemen energi. Sistem ini juga dilengkapi dengan integrasi Human-Machine Interface (HMI) untuk memantau, mengubah data, dan mengontrol sistem secara fleksibel. Hasil menunjukkan sistem mampu mengoptimalkan penggunaan panel surya dan BESS berdasarkan prediksi iradiasi matahari dan prioritas beban.
==============================================================================================================================
Electrical energy is a vital element in building operations, accounting for 30-45% of global energy consumption. Smart buildings utilize renewable energy sources such as solar panels; however, their performance is highly weather-dependent, necessitating energy storage systems (BESS) to maintain power supply stability. This study develops an energy management system based on the Mixed Integer Linear Programming (MILP) algorithm and employs XGBoost for energy demand forecasting (load forecasting) and solar panel output forecasting. Accurate predictions enable the optimal utilization of solar panels and BESS, reducing reliance on the power grid. The system was tested at Chamchuri 5 Building, Chulalongkorn University, Thailand, and the Department of Electrical Engineering, ITS Surabaya, Indonesia. At Chamchuri 5, the solar panel output forecast error reached an MSE of 302.47 Watts, and the load forecast error was 2633.31 Watts. At ITS, the solar panel output forecast error was 234.37 Watts, and the load forecast error was 1721 Watts. Additionally, the Wireless Power Meter used in the system exhibited a voltage error of 10.53 Volts and a current error of 0.03 Amps. The energy management system also integrates a smart switch as a controllable hardware device and a Human-Machine Interface (HMI) for flexible monitoring, data modification, and system control. The results demonstrate the system's capability to optimize the use of solar panels and BESS based on solar irradiance forecasts and load prioritization.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | Manajemen Energi, Optimasi, Smart Building, Renewable Energy, XGBoost, Artificial Intelegence |
Subjects: | T Technology > T Technology (General) T Technology > TK Electrical engineering. Electronics Nuclear engineering Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4050 Electronic information resources |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis |
Depositing User: | Fysna Candra Pratama |
Date Deposited: | 01 Feb 2025 13:27 |
Last Modified: | 01 Feb 2025 13:27 |
URI: | http://repository.its.ac.id/id/eprint/117457 |
Actions (login required)
![]() |
View Item |