Classifying Tomorrow’s Currents: A Probabilistic Neural Network Approach to Forecasting Electricity Prices

This study introduces a probabilistic neural network model for classifying next-day electricity prices into negative or positive categories. It addresses the challenge of price volatility exacerbated by renewable energy integration, presenting significant risks for market participants. Through a comparative analysis with the traditional support vector machine with radial basis function kernel using the German Day-Ahead market data of 2020, the probabilistic neural network demonstrates superior accuracy and yielded promising electricity price forecasting results. This approach equips market participants with valuable insights for strategic planning and risk mitigation.