Prediksi Produksi Daya Listrik Pada Panel Surya Berbahan Monokristalin Berdasarkan Kondisi Lingkungan Menggunakan Jaringan Saraf Tiruan Di Daerah Cepu, Jawa Tengah

Nugraha, Azizah Laksmi (2023) Prediksi Produksi Daya Listrik Pada Panel Surya Berbahan Monokristalin Berdasarkan Kondisi Lingkungan Menggunakan Jaringan Saraf Tiruan Di Daerah Cepu, Jawa Tengah. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 02311940000168-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (4MB) | Request a copy

Abstract

Prediksi produksi daya listrik pada panel surya dibutuhkan untuk mengetahui operasi dari system produksi daya listrik panel surya. Namun, akibat ketidakstabilan, intermiten, dan bentuk acak dari energi surya, prediksi produksi daya listrik pada panel surya semakin diperlukan. Pada penelitian ini, dicari juga hubungan antara variabel meteorologi dengan variabel keluaran dari panel surya menggunakan koefisien korelasi Pearson’s dan digunakan Convolutional Neural Network (CNN) untuk memprediksi produksi daya listrik pada panel surya berbahan monokristalin. Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), dan Root Mean Square Error (RMSE) digunakan untuk mengevaluasi performansi model prediksi dari penelitian ini. Dari hasil yang didapat, nilai koefisien korelasi terbesar pada hubungan antara suhu lingkungan dengan tegangan sebesar 0,5754 pada data kondisi float. Nilai koefisien korelasi radiasi dengan arus meningkat dengan data kondisi float dibandingkan dengan data keseluruhan disebabkan arus dipengaruhi oleh kondisi baterai, semakin penuh baterai semakin kecil arus mengalir menyebabkan penurunan efisiensi panel. Hasil perancangan sistem prediksi daya keluaran PV sudah mendekati daya aktualnya, dengan nilai performa RMSE 0,0537. Sedangkan CNN telah sesuai memprediksi daya dari panel, dengan nilai performa MAPE adalah 18.7633%. MAE sebesar 0.0176 dan RMSE sebesar 0.0466 sehingga sistem prediksi diartikan sudah baik dalam melakukan prediksi.
=======================================================================================================================================
Prediction of electric power production in solar panels is needed to determine the operation of the solar panel electric power production system. However, due to the unstable, intermittent, and random forms of solar energy, predicting the production of electric power in solar panels is increasingly needed. In this study, the relationship between meteorological variables and output variables from solar panels was sought using Pearson's correlation coefficient and Convolutional Neural Network (CNN) to predict the production of electric power in monocrystalline solar panels. Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Square Error (RMSE) are used to evaluate the performance of the prediction model of this study. From the results obtained, the largest correlation coefficient value is in the relationship between ambient temperature and voltage of 0.5754 in float condition data. The value of the correlation coefficient of radiation with current increases with the float condition data compared to the overall data because the current is affected by the battery condition, the fuller the battery the smaller the current flows causing a decrease in panel efficiency. The results of the PV output power prediction system design are close to the actual power, with a RMSE performance value of 0.0537. Meanwhile, CNN has correctly predicted the power from the panel, with a MAPE performance value of 18.7633%. MAE is 0.0176 and RMSE is 0.0466 so that the prediction system is interpreted to be good at making predictions.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	CNN, CNN, Koefisien Korelasi, Correlation Coefficient, Panel Surya, Solar Panels, Prediksi Daya Listrik, Electrical Power Prediction
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1087 Photovoltaic power generation
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User:	Azizah Laksmi Nugraha
Date Deposited:	23 Jul 2023 00:02
Last Modified:	23 Jul 2023 00:02
URI:	http://repository.its.ac.id/id/eprint/98893

Actions (login required)

View Item