Variable weight combined forecasting method-based electric power load short-term forecasting method

Abstract

The present invention discloses a variable weight combined forecasting method-based electric power load short-term forecasting method. According to the present invention, aiming at the disadvantages that a conventional single load forecasting model and a fixed weight combined forecasting model are low in forecasting accuracy, and by comprehensively considering the time correlation of the electricpower loads and the influence of other related factors on the electric power loads, and combining various electric power load forecasting models, a variable weight combined forecasting model of the electric power load short-term forecasting is established. Meanwhile, aiming at the disadvantage that a particle swarm optimization variable weight parameter combined forecasting method traps in a localoptimal solution easily, a parameter dynamic adjusting particle swarm optimization algorithm is established, thereby realizing the optimization solution of the weight parameters of the variable weight combined forecasting model, and finally realizing the short-term forecasting of the electric power loads. The combined forecasting model of the present invention is better than a conventional load forecasting method, a fixed weight combined forecasting method and a particle swarm optimization variable weight parameter combined forecasting method, and has the higher forecasting accuracy.

Description

technical field [0001] The invention belongs to the technical field of electric power system load forecasting, in particular to a short-term electric load forecasting method based on a variable weight combined forecasting method. Background technique [0002] Power system scheduling is the key to ensure the reliability and safety of the power system, and short-term power load forecasting is the basis of power system scheduling. Effective and accurate forecasting of short-term power load can effectively improve the security and economy of the power grid. [0003] Short-term and ultra-short-term load forecasting are generally aimed at forecasting hours and below, and are mainly used for power system dispatching. Due to its short forecast period, the forecast method is required to have a faster forecast time. In addition, short-term forecasting and ultra-short-term forecasting also require high accuracy. [0004] With the rapid development of my country's economy, the power ...

AI Technical Summary

Problems solved by technology

Fixed weights have significant disadvantages, because as the training samples are updated, the method cannot dynamically adjust the weights of each method, which eventually reduces the prediction accuracy

Images