Comparative Analysis of ANN and LSTM Models for Photovoltaic Panel Temperature Prediction in Hot Climates
Bibliographic record
Abstract
The temperature forecast of photovoltaic (PV) panels plays a major role in optimizing their performance and efficiency, particularly in hot regions like Marrakech, where accurate predictions can enhance energy production management and reduce kWh costs.This study evaluates the reliability of artificial neural networks (ANN) and long short-term memory networks (LSTM) for modeling PV panel temperature profiles, providing critical insights before system installation.Using root mean square error (RMSE) and coefficient of determination (R ) as key metrics, we compare the predictive performance of both models.The ANN model demonstrates marginally better accuracy in temperature prediction, with a training RMSE of 1.7231 (R = 0.974) and a test RMSE of 1.952 (R = 0.970).In contrast, the LSTM model shows slightly higher training and test RMSE values (1.8908 and 1.959, respectively) but maintains competitive R scores (0.970 and 0.969).While ANN exhibits a slight edge in RMSE, LSTM demonstrates greater stability in test loss over time, suggesting that its performance may be more robust in certain operational contexts.The minimal difference in R values (0.970 vs. 0.969) further indicates near-identical predictive capability between the two models.The choice of RMSE over mean absolute error (MAE) is justified by its sensitivity to larger deviations, which is critical for identifying extreme temperature fluctuations that could impact PV efficiency.Overall, both models generalize effectively, with ANN providing marginally better precision in our case study, while LSTM offers potential advantages in long-term stability.These findings highlight the importance of model selection based on specific operational requirements, whether prioritizing immediate prediction accuracy or long-term consistency in PV temperature forecasting.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".