Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
March 29, 2026EnergiesOpen Access

A CEEMDAN-CNN-BiLSTM-SDQN Framework for Photovoltaic Power Forecasting: Integrating Multi-Scale Decomposition with Adaptive Reinforcement Learning Compensation

View Full Paper
Ask AI
Bookmark
Share

Authors

WJWeijie JiaNorthwest A&F UniversityKLKeying LiuNorthwest A&F UniversityJXJiayi XuZhejiang International Studies University

Discussion

Loading...

Member takes

Implication

This framework integrates multi-scale decomposition and adaptive reinforcement learning to improve power forecasting accuracy in PV systems, suggesting enhanced reliability for renewable integration.

Key Points

  • The aim is to develop an accurate photovoltaic power forecasting method that overcomes the limitations of traditional approaches.
  • Integrates CEEMDAN for multi-scale decomposition of PV power series.
  • Employs parallel CNN-BiLSTM models to extract features from subcomponents.
  • Introduces an SDQN agent for real-time error compensation.
  • Achieves RMSE of 0.4463, MAE of 0.1256, MAPE of 1.2814%, and R2 of 92.58%.
  • Outperforms benchmark models in forecasting accuracy.
  • CEEMDAN significantly reduces mode mixing and enhances prediction precision.

Cite This Study

Jia et al. (2026) studied this question.

synapsesocial.com/papers/69c8c277de0f0f753b39cc6fhttps://doi.org/10.3390/en19071649
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Short-term power prediction of photovoltaic power stations based on Kepler optimization algorithm and VMD-CNN-LSTM model2025 · 6 citations
  2. 2Photovoltaic Short-Term Output Power Forecast Model Based on Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise–Kernel Principal Component Analysis–Long Short-Term Memory2024 · 14 citations
  3. 3Prediction of Ultra-Short-Term Photovoltaic Power Using BiLSTM–Informer Based on Secondary Decomposition2025 · 9 citations
  4. 4Robust multi-scale LSTM for reliable and intelligent photovoltaic power forecasting2026 · 4 citations