Aprilyan, Verdiansyah Ayus (2026) Visualization of Minute-Level Solar Radiation Prediction in Medan City Using a CNN–GRU Model. Undergraduate thesis, UPN Veteran Jawa Timur.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
22081010015.-cover.pdf Download (555kB) |
|
Text (Bab 1)
22081010015.-bab1.pdf Download (190kB) |
|
Text (Bab 2)
22081010015.-bab2.pdf Restricted to Repository staff only until 17 July 2028. Download (531kB) |
|
Text (Bab 3)
22081010015.-bab3.pdf Restricted to Repository staff only until 17 July 2028. Download (555kB) |
|
Text (Bab 4)
22081010015.-bab4.pdf Restricted to Repository staff only until 17 July 2028. Download (3MB) |
|
Text (Bab 5)
22081010015.-bab5.pdf Download (107kB) |
|
Text (Daftar pustaka)
22081010015.-daftarpustaka.pdf Download (196kB) |
|
Text (Lampiran)
22081010015.-lampiran.pdf Restricted to Repository staff only Download (132kB) |
Abstract
This study develops a hybrid CNN-GRU model to predict per-minute solar radiation in Medan City to support the operational efficiency of Solar Power Plants (PLTS). The proposed model integrates three 1D CNN layers with 64, 32, and 16 filters to extract spatial features and two GRU layers with 50 and 32 units to capture longterm temporal dependencies in time-series data. The dataset consists of local meteorological data from Medan City with seven main features, processed through data filtering, missing value interpolation, smoothing, MinMaxScaler normalization, and sliding window sequencing. Systematic hyperparameter tuning identified the optimal configuration with a window size of 30, a learning rate of 0.001, a batch size of 512, and a data split ratio of 90% for training, 5% for validation, and 5% for testing. The final model achieved high prediction performance with an R² of 0.9923, an RMSE of 12.9430, an MAE of 6.3709, and a MAPE of 8.50%, demonstrating its effectiveness in accurately predicting minutelevel solar radiation. The prediction results were further visualized through an interactive web-based system using Folium and TimestampedGeoJson, providing a spatio-temporal digital map of Medan City with per-minute animation, color gradients representing solar radiation intensity, and daily and monthly prediction dashboards.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QA Mathematics > QA76.6 Computer Programming |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Verdiansyah Ayus Aprilyan | ||||||||||||
| Date Deposited: | 17 Jul 2026 07:27 | ||||||||||||
| Last Modified: | 17 Jul 2026 07:27 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/55738 |
Actions (login required)
 |
View Item |