基于遗传灰色神经网络模型的实时电价条件下短期电力负荷预测

A Gray Neural Network Model Improved by Genetic Algorithm for Short-Term Load Forecasting in Price-Sensitive Environment

在智能电网条件下,用户的用电模式将会发生重大变化,其中一个显著的变化就是用户可以根据电能需求结合实时电价调整其消费模式。这使得用户负荷预测更为复杂。在对影响短期电力负荷特性的各种因素进行分析的基础上,综合考虑了实时电价的影响,提出了一种用遗传算法优化改进的灰色神经网络方法,利用灰色模型可以弱化数据的随机性以及神经网络的高度非线性,对短期负荷进行预测,采用遗传算法对网络进行优化,从而提高了预测的精确度。实例证明该算法能较好地解决实时电价下的短期负荷预测问题。

In the condition of smart grid, there will be many great changes in power consumption pattern of power consumers, and a significant change in the power consumption pattern is that based on the real-time price consumers can adjust their electricity consumption modes and it makes the forecasting of power load complicated. On the basis of analyzing various factors impacting on short-term load characteristics, based on the synthetical consideration of real-time price a gray neural network method improved by genetic algorithm is proposed. Utilizing the property of gray model that the randomness of data can be reduced by it and the strong nonlinearity of neural network the short-term load forecasting is performed and the genetic algorithm is used to optimize the neural network to improve the accuracy of forecasting. Results of computation example show that the proposed algorithm is available to short-term load forecasting under real-time price.