MUHAMMAD CALVIN DWI SAPUTRA, NIM. 132021077 (2025) PREDIKSI DAYA PEMBANGKIT LISTRIK TENAGA SURYA MENGGUNAKAN METODE K-NEAREST NEIGHBOR. Skripsi thesis, Universitas Muhammadiyah Palembang.

Preview	Text 132021077_BAB I_DAFTAR PUSTAKA.pdf - Published Version Download (1MB) | Preview
	Text 132021077_BAB II.pdf Restricted to Repository staff only Download (633kB)
	Text 132021077_BAB III.pdf Restricted to Repository staff only Download (327kB)
	Text 132021077_BAB IV.pdf Restricted to Repository staff only Download (916kB)
	Text 132021077_BAB V.pdf Restricted to Repository staff only Download (250kB)
	Text 132021077_DAFTAR PUSTAKA.pdf Restricted to Repository staff only Download (237kB)
	Text 132021077_LAMPIRAN.PDF.pdf Restricted to Repository staff only Download (934kB)
	Text 132021077_Cover_Sampai_Lampiran.pdf Restricted to Repository staff only Download (2MB)

Abstract

abstrak Penelitian ini membahas prediksi daya keluaran PLTS dengan memanfaatkan parameter suhu udara, kelembaban, intensitas cahaya, dan kecepatan angin. Metode K-Nearest Neighbor (KNN) digunakan sebagai model utama, kemudian dibandingkan dengan Support Vector Regression (SVR) dan Multi-Layer Perceptron (MLP). Dataset berjumlah 250 sampel dengan pembagian 80% untuk pelatihan dan 20% untuk pengujian. Hasil analisis menunjukkan bahwa intensitas cahaya memiliki pengaruh paling dominan terhadap daya PLTS, sedangkan parameter lain berperan sebagai pendukung. Evaluasi model memperlihatkan KNN memiliki kinerja terbaik dengan nilai R² = 0.616 dan RMSE = 28.28, diikuti SVR (R² = 0.605) dan MLP (R² = 0.289). Dengan demikian, KNN dapat dianggap sebagai metode yang lebih akurat dalam memprediksi daya keluaran PLTS pada dataset terbatas. Kata kunci: PLTS, Prediksi daya, K-Nearest Neighbor, RMSE, R² abstract This study focuses on predicting PLTS output power using environmental parameters such as air temperature, humidity, solar irradiance, and wind speed. The K-Nearest Neighbor (KNN) method was applied as the primary model and compared with Support Vector Regression (SVR) and Multi-Layer Perceptron (MLP). The dataset consisted of 250 samples, divided into 80% for training and 20% for testing. The results indicate that solar irradiance has the strongest influence on PLTS output power, while other parameters act as supporting factors. Model evaluation shows that KNN achieved the best performance with R² = 0.616 and RMSE = 28.28, outperforming SVR (R² = 0.605) and MLP (R² = 0.289). Therefore, KNN can be considered a more accurate and reliable method for predicting PLTS output power on limited datasets. Keywords: Solar Power Plant, Power prediction, K-Nearest Neighbor, RMSE, R²

Item Type:	Thesis (Skripsi)
Additional Information:	Pembimbing : 1. Dr. Bengawan Alfaresi, S.T., M.T 2. Dr. Feby Ardianto, S.T., M.Cs
Uncontrolled Keywords:	Kata kunci: PLTS, Prediksi daya, K-Nearest Neighbor, RMSE, R²
Subjects:	Elektro > Fisika Terapan Elektro > Pengujian dan Pengukuran Listrik Elektro > Teknik Listrik
Divisions:	Fakultas Teknik > Teknik Elektro (S1)
Depositing User:	Mahasiswa Fakultas Teknik
Date Deposited:	23 Oct 2025 05:40
Last Modified:	23 Oct 2025 05:40
URI:	http://repository.um-palembang.ac.id/id/eprint/33950

Actions (login required)

View Item