基于MEA-Elman神经网络的电力日负荷预测

摘要

Elman神经网络是一种动态反馈网络,对历史状态敏感,具有短期记忆功能和处理动态信息的能力,可以建立动态、非线性电力负荷预测模型.由于Elman神经网络采用BP算法,容易陷入局部极小解,迭代次数多且学习效率低,该文利用思维进化算法(MEA)优化Elman神经网络的方法,提出基于MEA-Elman神经网络的电力负荷预测模型.实验表明,该方法能够避免不成熟收敛问题,减少迭代次数,有效提高了配电网短期负荷的预测精度,对电力系统合理调度与规划具有重要意义.%As we know, Elman neural network is a kind of dynamic feedback network, which is sensitive to historical state.Elman has short-term memory function and ability to process dynamic information, and dynamic nonlinear power load forecasting model can be established with it.However, due to BP algorithm is used for Elman neural network, it is easy to fall into the local minimal solution, so the number of iterations is high while the learning efficiency is low.In this paper, daily power load forecasting model based on MEA-Elman neural network is presented using method of Elman neural network optimized by the mind evolutionary algorithm (MEA).Experiments show that this method can avoid the problem of premature convergence, reduce the number of iterations, and improve the prediction accuracy of short-term load of power distribution.It is significant for the reasonable scheduling and planning of power system.