Research And Implement On Short-term Load Forecasting Technology Of Electric Power System

Short-term load forecasting (STLF) is essential part of energy management system. It is not only the key basis for the distribution center to ensure operating safely and economically and to achieve scientific management in the grid, but also a necessary content of the electricity marketplace operation management. The prediction accuracy has direct influence on economy benefits of the grid and the power plants.The change of power system load is prone to be affected by many factors, especially effects of weather and vacation, which common methods of load forecasting can not be adapted well. After analyzing the advantage and disadvantage of common methods, writer put forward a new pattern of Back-Propagation neural network, which concerns with the factor of weather and type of date and details in data process. Finally, the article designed and implemented the pattern by Object-Oriented programming language and analyzed the performance through the experiment.