Artificial Intelligence-driven Algorithms for Precision Forecasting in Electricity Pricing

Abstract

Electricity price forecasting is essential for ensuring the efficient and cost-effective operation of the power industry. Accurate and precise prediction of electricity prices is critical in the electricity market due to the inherent challenges posed by factors such as high volatility, seasonality, calendar effects, and non-linearity. This paper presents four deep learningbased models—Multilayer Perceptron (MLP), Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and a hybrid Convolutional Neural Network-Long Short-Term Memory (CNNLSTM)—to forecast electricity prices in the Nord Pool Spot Market. The approach utilizes historical electricity price data, making it a univariate time-series forecasting method. The performance of these models is assessed using various accuracy metrics and visualized through box plots. The simulation results indicate that the CNN-LSTM model consistently outperforms the other three models across most evaluation metrics.