Enhancing photovoltaic energy output predictions using ANN and DNN: a hyperparameter optimization approach
| dc.contributor.author | Coşgun, Atıl Emre | |
| dc.date.accessioned | 2025-10-15T11:13:50Z | |
| dc.date.available | 2025-10-15T11:13:50Z | |
| dc.date.issued | 2025 | |
| dc.department | Mühendislik Fakültesi | |
| dc.description.abstract | This study investigates the use of artificial neural networks (ANNs) and deep neural networks (DNNs) for estimating photovoltaic (PV) energy output, with a particular focus on hyperparameter tuning. Supervised regression for photovoltaic (PV) direct current power prediction was conducted using only sensor-based inputs (PanelTemp, Irradiance, AmbientTemp, Humidity), together with physically motivated-derived features (ΔT, IrradianceEff, IrradianceSq, Irradiance × ΔT). Samples acquired under very low irradiance (<50 W m−2) were excluded. Predictors were standardized with training-set statistics (z-score), and the target variable was modeled in log space to stabilize variance. A shallow artificial neural network (ANN; single hidden layer, widths {4–32}) was compared with deeper multilayer perceptrons (DNN; stacks {16 8}, {32 16}, {64 32}, {128 64}, {128 64 32}). Hyperparameters were selected with a grid search using validation mean squared error in log space with early stopping; Bayesian optimization was additionally applied to the ANN. Final models were retrained and evaluated on a held-out test set after inverse transformation to watts. Test performance was obtained as MSE, RMSE, MAE, R2, and MAPE for the ANN and DNN. Hence, superiority in absolute/squared error and explained variance was exhibited by the ANN, whereas lower relative error was achieved by the DNN with a marginal MAE advantage. Ablation studies showed that moderate depth can be beneficial (e.g., two-layer variants), and a simple bootstrap ensemble improved robustness. In summary, the ANN demonstrated superior performance in terms of absolute-error accuracy, whereas the DNN exhibited better consistency with relative-error accuracy. | |
| dc.identifier.doi | 10.3390/en18174564 | |
| dc.identifier.issn | 19961073 | |
| dc.identifier.issue | 17 | |
| dc.identifier.scopus | 2-s2.0-105015968732 | |
| dc.identifier.scopusquality | Q1 | |
| dc.identifier.uri | https://doi.org/10.3390/en18174564 | |
| dc.identifier.uri | https://hdl.handle.net/20.500.12451/14815 | |
| dc.identifier.volume | 18 | |
| dc.indekslendigikaynak | Scopus | |
| dc.institutionauthor | Coşgun, Atıl Emre | |
| dc.institutionauthorid | 0000-0002-4889-300X | |
| dc.language.iso | en | |
| dc.publisher | Multidisciplinary Digital Publishing Institute (MDPI) | |
| dc.relation.ispartof | Energies | |
| dc.relation.publicationcategory | Makale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı | |
| dc.rights | info:eu-repo/semantics/openAccess | |
| dc.subject | ANN | |
| dc.subject | DNN | |
| dc.subject | Epoch | |
| dc.subject | Hidden Layer | |
| dc.subject | Hyperparameter | |
| dc.subject | Learning Rate | |
| dc.title | Enhancing photovoltaic energy output predictions using ANN and DNN: a hyperparameter optimization approach | |
| dc.type | Article |
