Model Oriented to Electricity Spot Market Clearing Price Forecasting

As renewable energy becomes widespread and power markets advance rapidly, clearing prices in electricity spot markets display high volatility, nonlinearity, and uncertainty. Accurately forecasting these prices is vital for the healthy operation and economic benefits of power markets. This paper presents a prediction model for electricity spot market clearing prices based on a Parallel Temporal Convolutional Network (P-TCN). The P-TCN model integrates multi - period features, using independent TCN structures to model different - period features like hours, days, weeks, and months. It uses dilated convolutions to expand the receptive field, effectively solving training complexity and gradient vanishing issues in traditional CNNs when dealing with long sequences. Experimental results show that this model can enhance the prediction accuracy of electricity spot market clearing prices, offering strong support for benefit optimization and stable operation in power markets.