Publication

Abstract : Energy market planning and decision-making heavily rely on electricity price predictions. Energy market players may manage the risks associated with electricity price volatility, maximize energy production and consumption, and make educated decisions about their investment plans with the aid of accurate and trustworthy electricity price forecasts. A mixed neural network model on Bi-LSTM-GRU architecture is presented in this article for predicting power prices. The Bidirectional Long Short-Term Memory (Bi-LSTM) and Gated Recurrent Unit (GRU) are two well-known neural network architectures. The suggested model combines their characteristics to increase the accuracy and resilience of power price predictions. Using publicly accessible energy pricing statistics, the effectiveness of the suggested model is assessed and contrasted with leading models. The findings show that the Bi-LSTM-GRU hybrid model beats the current models in terms of specificity and robustness of prediction. For players in the energy market, regulators, and policymakers, our suggested model can offer insightful information and help with decision-making.