Research on Combined Ultra-short-term PV Power Prediction Method Based on VMD-WOA-BiLSTM

Photovoltaic (PV) power generation offers significant environmental benefits and is one of the effective means to address environmental issues fundamentally. In the operation of PV power plants, accurate power prediction technology is crucial for daily scheduling and power generation planning. To address the challenges of high randomness and low prediction accuracy in PV power forecasting, this study proposes a PV power prediction model based on Variational Mode Decomposition (VMD), Whale Optimization Algorithm (WOA), and Bidirectional Long Short-Term Memory (BiLSTM) neural networks. To tackle the issue of high volatility in PV power output data, the VMD data decomposition algorithm is employed to decompose the original power data, thereby enhancing the reliability of the raw data and reducing the impact of noise in the dataset. Additionally, to overcome the problems of large prediction biases and low accuracy caused by improper selection of hyperparameters in traditional prediction models, the WOA is utilized to optimize the hyperparameters of the BiLSTM prediction model, thus improving the reliability of the prediction model. Finally, the proposed model is validated through case simulations and compared with control groups. The results demonstrate the practicality and effectiveness of the proposed method.