Visualisasi Nilai Ekonomis Panel Surya Menggunakan Metode Exponential Smoothing dan Feedforward Backpropagation

Akmal, Dara Nasywa Fathya Afiqah (2024) Visualisasi Nilai Ekonomis Panel Surya Menggunakan Metode Exponential Smoothing dan Feedforward Backpropagation. Other thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5026201074-Undergraduate_Thesis.pdf] Text
5026201074-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2026.

Download (4MB) | Request a copy

Abstract

Peningkatan konsumsi listrik di Indonesia saat ini didominasi oleh sumber energi fosil. Sementara potensi energi terbarukan berupa sinar matahari sangat melimpah, tetapi pemanfaatannya masih minim dan tercatat baru menacapai sekitar 0.08%. Salah satu solusi strategis yang dapat diadopsi adalah dengan pemasangan panel surya fotovoltaik (PV) pada atap bangunan skala rumah tangga. Namun, permasalahan utama pemanfaatan PV adalah fluktuasi produktivitas energi listrik yang bergantung pada variabilitas cuaca setempat. Penelitian ini bertujuan untuk mengembangkan model ramalan daya keluaran PV agar dapat diaplikasikan untuk proyeksi dan evaluasi potensi penghematan pemakaian listrik pada rumah tangga pengguna panel surya. Berdasarkan hasil penelitian pada tiga lokasi, keterkaitan antara variabel cuaca dan produksi energi bervariasi untuk setiap daerah di mana daerah Malang dipengaruhi secara kuat oleh suhu minimal dan kelembapan relatif, daerah Jakarta dipengaruhi oleh suhu rata-rata, dan daerah Surabaya dipengaruhi oleh suhu maksimal. Adapun digunakan dua tahap pembelajaran pada penelitian ini berupa peramalan cuaca menggunakan Exponential Smoothing dan prediksi luaran energi menggunakan Artificial Neural Networks. Dengan menggunakan MAPE sebagai salah satu metrik evaluasi, performa prediksi energi di setiap daerah tersebut berturut-turut mencapai 16.49%, 19.46%, serta 21.06% dengan total penghematan selama tujuh hari kedepan mencapai kisaran Rp31,474.31 – Rp42,877.52.
=================================================================
The increase in Indonesia’s electricity consumption is currently dominated by fossil energy sources. While the potential for renewable energy such sunlight is very abundant, its utilization is still minimal and has only reached around 0.08%. The strategic solution that can be adopted is to install photovoltaic (PV) solar panels on the roofs of household-scale buildings. However, the main problem with PV utilization is fluctuations in electrical energy productivity which depend on local weather variability. This research aims to develop a PV output power forecasting model so that it can be applied to project and evaluate the potential for saving electricity usage in households using solar panels. Based on research results in three locations, the relationship between weather variables and energy production varies for each region, where the Malang area is strongly influenced by minimum temperature and relative humidity, the Jakarta area is influenced by average temperature, and the Surabaya area is influenced by maximum temperature. Two stages of learning were used in this research, namely weather forecasting using Exponential Smoothing and energy output prediction using Artificial Neural Networks. By using MAPE as one of the evaluation metrics, the energy prediction performance in each of these areas respectively reached 16.49%, 19.46%, and 21.06% with total savings over the next seven days reaching the range of IDR 31,474.31 – IDR 42,877.52.

Item Type: Thesis (Other)
Uncontrolled Keywords: Energi Terbarukan, Exponential Smoothing, Feedforward Backpropagation, Panel Surya, Prediksi Energi Listrik. ================= Exponential Smoothing, Feedforward Backpropagation, Solar Panel, Power Output Prediction, Renewable Energy.
Subjects: T Technology > T Technology (General) > T174 Technological forecasting
T Technology > T Technology (General) > T385 Visualization--Technique
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1001 Production of electric energy or power
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1087 Photovoltaic power generation
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1322.6 Electric power-plants
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis
Depositing User: Dara Nasywa Fathya Afiqah Akmal
Date Deposited: 02 Aug 2024 05:48
Last Modified: 02 Aug 2024 05:48
URI: http://repository.its.ac.id/id/eprint/112256

Actions (login required)

View Item View Item