Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/29705| Title: | Rancang Bangun Sistem Tracker dan Wiper Panel Surya Menggunakan Metode Articial Intelligence |
| Other Titles: | Design and Construction of a Solar Panel Tracker and Wiper System Using Artificial Intelligence Methods |
| Authors: | Prasetya, Ragil |
| metadata.dc.contributor.advisor: | Satria, Habib |
| Keywords: | Solar panel;automatic tracking;automatic cleaning;Arduino;Artificial Intelligence;Windy;SunCalc;pelacak otomatis;Panel surya;pembersih otomatis |
| Issue Date: | Sep-2025 |
| Publisher: | Universitas Medan Area |
| Series/Report no.: | NPM;218120003 |
| Abstract: | Penelitian ini membahas perancangan dan pengembangan sistem pelacak dan pembersih otomatis pada panel surya dengan memanfaatkan metode Artificial Intelligence (AI) sebagai acuan dalam menentukan sudut optimal terhadap posisi matahari. Sistem ini dirancang untuk meningkatkan efisiensi konversi energi dengan mengatur sudut orientasi panel secara otomatis serta menjaga kebersihan permukaan panel melalui kontrol terpadu. Sistem pelacak menggunakan sensor LDR untuk mendeteksi intensitas cahaya dari dua arah, yang kemudian menggerakkan aktuator agar panel mengikuti pergerakan matahari. Sistem pembersih otomatis bekerja berdasarkan pembacaan sensor debu, dan motor pembersih akan diaktifkan apabila ambang batas kotoran terlampaui. Penerapan AI dilakukan secara tidak langsung dengan memanfaatkan data cuaca dan posisi matahari dari platform Windy dan SunCalc sebagai acuan referensi dan evaluasi kinerja sistem. Hasil pengujian menunjukkan bahwa sistem pelacak mampu menyesuaikan sudut panel dengan deviasi rata-rata sebesar ±5–10° dibandingkan dengan data referensi AI. Sementara itu, sistem pembersih terbukti efektif dalam meningkatkan tegangan dan daya keluaran setelah proses pembersihan dilakukan. Dengan demikian, integrasi sistem otomatis berbasis sensor dengan dukungan data AI dari Windy dan SunCalc mampu meningkatkan kinerja dan efisiensi panel surya dalam memanfaatkan energi terbarukan secara optimal. This study presents the design and development of an automatic solar panel tracking and cleaning system, utilizing Artificial Intelligence (AI) as a reference for determining the optimal solar angle. The system aims to improve energy conversion efficiency by automatically adjusting the panel’s orientation and maintaining its surface cleanliness through an integrated control mechanism. The tracking component uses Light Dependent Resistors (LDRs) to detect light intensity from two directions, which drives the actuator to follow the sun’s movement. The cleaning system operates based on a dust sensor that activates a cleaning motor once dust accumulation exceeds a predefined threshold. AI is implemented indirectly by referring to real-time environmental data sourced from the Windy and SunCalc platforms, which provide solar radiation and sun angle information to evaluate system performance. Testing results show that the tracker successfully aligns the panel with the sun, with an average angular deviation of ±5–10° compared to the AIbased reference angles. The cleaning system also contributes to higher voltage and power output after each cleaning cycle, confirming its effectiveness. In conclusion, the integration of sensor-based automation with AI-supported data from Windy and SunCalc enhances the adaptability and efficiency of solar panel systems in capturing renewable energy more optimally. |
| Description: | 78 Halaman |
| URI: | https://repositori.uma.ac.id/handle/123456789/29705 |
| Appears in Collections: | SP - Electrical Engineering |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| 218120003 - Ragil Prasetya - Chapter IV.pdf Restricted Access | Chapter IV | 1.86 MB | Adobe PDF | View/Open Request a copy |
| 218120003 - Ragil Prasetya - Fulltext.pdf | Cover, Abstract, Chapter I, II, III, V, Bibliography | 921.16 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.