Abstract:Accurate anticipation of photovoltaic (PV) power generation in advance is crucial for renewable energy development, infrastructure planning, efficient power grid operations, and energy management. Emerging data-driven methods represented by deep learning have provided effective solutions for PV power generation forecasting. However, conventional data-driven forecasting algorithms rely heavily on extensive historical data, making it challenging to predict the output of newly-built PV plants (NPP) due to limited data availability. To address this concern, a novel physics-infused transfer learning model is proposed for short-term cross-plant PV power prediction, which leverages the public prediction knowledge learnt from relevant PV plants to develop a predictor compatible with NPP. Key innovations include: (1) Firstly, we propose a forecasting-oriented Domain Adversarial Neural Network (DANN) that incorporates Wasserstein distance, enabling the reduction of the discrepancy in the feature vector distribution between source and target plants through iterative adversarial training. The feature vectors from NPP would be compatible with a predictor trained on the feature vectors from data-rich PV plants. (2) Secondly, this framework implicitly transfers the intricate regression patterns from data-rich PV plants to NPP with limited historical measurements, allowing for the capture of short-term fluctuations using real-time data as input. (3) Subsequently, a well-calibrated physical model chain enables the refined and stabilized numerical calculations in the conversion from solar irradiance to PV power output, thus extending the forecasting horizon. (4) Finally, the Bayesian combination model (BCM) is deployed to coordinate the two sub-predictors, enabling simultaneous prediction of trends and short-term fluctuations. The hybrid framework is formulated and validated adopting real-world PV data from four PV plants in North China. Simulation results indicate that the prediction accuracy outperforms all selected benchmark models. Compared to standalone models trained on the scarce data of NPP, the transfer learning-based model can reduce prediction errors by approximately 20% to 68%. Moreover, the proposed model demonstrates excellent generalizability and robustness, effectively mitigating the effects of geographical disparities among source and target domains, the performance of cross-geographic region transfer forecasting task is improved by 30%- 40% compared to the traditional alternative.

A Day-Ahead Photovoltaic Power Prediction via Transfer Learning and Deep Neural Networks

Transfer Learning for Photovoltaic Power Forecasting with Long Short-Term Memory Neural Network

Short-Term Photovoltaic Prediction Based on Transfer Learning and LSTM

Enhanced Photovoltaic Power Generation Forecasting for Newly-Built Plants Via Physics-Infused Transfer Learning with Domain Adversarial Neural Networks

Ultra-short-term PV power forecasting based on LSTM with PeepHoles connections

TransOS-ELM: A Short-term Photovoltaic Power Forecasting Method Based on Transferred Knowledge from Similar Days

Photovoltaic power forecasting: A hybrid deep learning model incorporating transfer learning strategy

Prediction of photovoltaic power generation based on LSTM and transfer learning digital twin

Deep and Machine Learning Models to Forecast Photovoltaic Power Generation

An adaptive deep learning framework for day-ahead forecasting of photovoltaic power generation

Deep Attention ConvLSTM-based Adaptive Fusion of Clear-Sky Physical Prior Knowledge and Multivariable Historical Information for Probabilistic Prediction of Photovoltaic Power

A Hybrid Photovoltaic Power Prediction Model Based on Multi-source Data Fusion and Deep Learning

Solar Radiation Prediction in PV Power Systems: a Comparison of Deep Learning Models Using Big Data

Solar power generation prediction based on deep Learning

Distributed-regional photovoltaic power generation prediction with limited data: A robust autoregressive transfer learning method

Daily Prediction Model of Photovoltaic Power Generation Using a Hybrid Architecture of Recurrent Neural Networks and Shallow Neural Networks

Accurate Solar PV Power Prediction Interval Method Based on Frequency-Domain Decomposition and LSTM Model

Deep learning neural networks for short-term photovoltaic power forecasting

Short-term Photovoltaic Power Prediction Based on Daily Feature Matrix and Deep Neural Network

A hybrid machine learning forecasting model for photovoltaic power

A novel GBDT-BiLSTM hybrid model on improving day-ahead photovoltaic prediction