Electricity Price Forecasting Enhancement Using Combined Model

Electricity price forecasting is an important tool for market participants to design bidding strategies in the electricity market. This paper introduces a combined neural network (c-NN) model that integrates Temporal Convolutional Networks (TCNs), Long Short-Term Memory (LSTM) networks, CNN-GRU hybrids, and Transformer models to address the volatility and nonlinear patterns of electricity prices. Our evaluation uses historical data from ERCOT Hub North prices to demonstrate that this integrated approach significantly outperforms individual models in accuracy. The findings suggest substantial financial benefits and support for renewable energy transitions, offering valuable insights for energy traders, grid operators, and policy makers. This research underscores the potential of combining machine learning (ML) models to enhance energy market analytics.