基于Elman神经网络的电力负荷预测模型研究

Research on Power Load Forecasting Model Based on Elman Neural Network

针对传统的电力负荷预测的不足,利用Elman神经网络时变能力强、误差可控、预测精度高等优点,提出一种基于循环结构Elman神经网络的电力负荷预测模型。采用学校实习基地的某企业半个月的电力负荷历史数据作为原始数据,并通过MATLAB平台进行仿真,对提出的预测方法和模型进行了研究,分析检验所建立模型的实际预测能力。首先对负荷样本数据归一化运算,使Elman神经网络输入层变量集中在[0,1]区间内,再对所建立的网络模型进行训练,使系统具备适应时变能力,增强系统全局稳定性。仿真分析结果显示:基于Elman神经网络的电力负荷预测模型预测误差小、精度高,预测速度快,表明了所提出的预测模型具有一定的实用价值。

Aiming at the deficiency of traditional power load forecasting,a power load forecasting model based on cyclic structure Elman neural network is proposed by taking advantage of the advantages of strong time-varying ability,controllable error and high prediction accuracy.The historical power load data of an enterprise in the school practice base for half a month was used as the original data,and the MATLAB platform is used for simulation.The proposed prediction method and model were studied,and the actual prediction ability of the model was analyzed and tested.Before the prediction,the load sample data should be normalized and the input layer variables of Elman neural network should be concentrated in the interval of[0,1].Then,the established network model should be trained to enable the system to adapt to time changes and enhance the global stability of the system.