Research on Net Load Forecasting Considering the Influence of Correlation Factors

In the face of increasing renewable energy integration, accurately predicting net load, defined as the difference between load demand and renewable generation, has become crucial for power system operation. This paper presents a net load forecasting model that incorporates multiple correlation influencing factors such as weather conditions, wind power, photovoltaic generation, and economic indicators. By employing multiple regression analysis and long-sequence time series predicting, we analyze the impact of these correlated factors on net load. The proposed model integrates probabilistic scenario generation and uncertainty characterization to capture variations in load patterns effectively. Experimental results show that the model outperforms traditional methods, demonstrating improved accuracy in predicting 8760 -hour annual load curves. This model could support enhanced resource scheduling and renewable energy utilization, providing a valuable tool for grid operators managing high variability and uncertainty in renewable generation.