Research on a Short-Term Electricity Price Forecasting Model Based on SSA-ESN-BiLSTM

Against the backdrop of the increasing integration of renewable energy sources, the dynamic characteristics of the electricity market have become more complex, with a significant increase in electricity price volatility, posing higher demands for the accuracy of electricity price forecasting. Current neural network models all have certain limitations. To address this challenge, this paper proposes a hybrid model based on Echo State Networks (ESN) and Bidirectional Long Short-Term Memory (BiLSTM) for electricity market price forecasting. The model first uses ESN for feature extraction to capture the deep structure and complex patterns of the data. Then, a Bi-directional LSTM network model is designed, which utilizes its sequential learning ability for electricity price prediction. Additionally, the Sparrow Search Algorithm (SSA) is employed for parameter optimization. Through empirical analysis and comparative experiments, the effectiveness of the proposed hybrid model in enhancing the accuracy of electricity price forecasts has been validated, particularly when dealing with highly volatile datasets, demonstrating good robustness and predictive performance. Moreover, the model is adaptable to different market conditions, providing more precise decision support for participants in the electricity market.