Day-Ahead Tariff Prediction Method for Power Trading Market Based on IQR-RANSAC

As the domestic electricity spot market rapidly evolves, accurate electricity price forecasting is essential for market participants to develop trading strategies, assess risks, and optimize resources. However, day-ahead prices are influenced by supply-demand dynamics, weather, generation costs, policy changes, and participant behavior, leading to significant volatility and prediction challenges. This paper evaluates these influencing factors and proposes a short-term prediction method using the IQR-RANSAC algorithm, which enhances prediction accuracy and stability by removing outliers. Experimental results indicate that the polynomial regression model combined with IQR-RANSAC effectively forecasts day-ahead prices. Business implications are discussed.