Day-Ahead Solar Irradiance Forecasting Using CEEMDAN-PCC-BiLSTM Hybrid Model
Bibliographic record
Abstract
Short-term solar irradiation prediction is essential for smooth running of various industries, especially for management of flawless electricity generation and distribution.However, solar datasets are rapid, noisy, and non-linear, making standalone models such as LSTM struggle to extract meaningful patterns for GHI forecasting.The unidirectional nature of LSTM restricts learning to past dependencies only, limiting its ability to forecast sudden irradiance drops.Additionally, vanishing gradient issues in long historical data hinder ability of LSTM to capture complex temporal dependencies, leading to suboptimal forecasting performance.To address these challenges, the proposed model leverages deep learning techniques with a feature-refinement approach.Initially, multiple intrinsic mode functions (IMFs) are extracted from the solar irradiance data, each representing different feature sets.A selection criterion is then applied to retain only the most relevant IMFs, which are integrated to form the input feature set.This refined input is used to train a deep learning network, resulting in day-ahead solar irradiance forecasting model.The performance of the model is evaluated at three distinct locations with varying climatic conditions to test its performance consistency.In the proposed model, reduced error rates are observed, which reflects in lower values of MAPE (2.31%), RMSE (1.55 W/m² ), and MAE (1.41 W/m² ) as compared to the standalone LSTM model, which records MAPE (3.23%), RMSE (3.21 W/m² ), and MAE (1.97 W/m² ).Results demonstrate significant improvements over benchmarking method corresponding to more than 90% improvement across all statistical parameters.Also, consistent accuracy across diverse climates highlights robustness of the proposed model which makes it a reliable and versatile solution for practical forecasting applications.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
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 source (direct Gemma or distilled Codex), 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".