Power system short-term load prediction method based on neural network containing dendritic structure

The invention relates to an electric power system short-term load prediction method based on a neural network containing a dendritic structure. The method belongs to the technical field of power system load prediction. An innovative power system short-term load prediction model is provided. Compared with a traditional neural network model, the method is innovative in that a calculation process of a dendritic structure is introduced, so that the processing capacity of the network is enhanced. According to the method, the prediction accuracy can be remarkably improved. The accurate load prediction not only can optimize the scheduling of the generator set, improve the power generation efficiency and ensure the economical efficiency, but also can promote the system to absorb renewable energy more effectively under the current large-scale renewable energy grid connection condition. 本发明为一种基于含树突结构神经网络的电力系统短期负荷预测方法。属于电力系统负荷预测的技术领域。提出了一种创新的电力系统短期负荷预测模型。相较于传统的神经网络模型，本方法的创新性在于引入了树突结构的计算过程，以此增强网络的处理能力。本方法能显著提升预测的准确性。精准的负