基于小波变换的风电上网电价的预测

提出基于小波变换和神经网络的预测模式,首先利用小波变换将历史销售电价序列分解为高频和低频序列,并分别构造学习样本作为神经网络的输入,对不同频率的序列分别采用神经网络进行预测,然后将不同频率预测结果通过小波重构,得到销售电价,并根据合理的输配电价管制模型推算上网电价,阐述影响风电上网电价的因素,实现对风电上网电价的预测。结果表明：提出的预测方法对美国PJM电力市场的历史节点边际电价（LMP）进行预测是有效的,从电力市场的角度入手分析风电上网电价机制具有重要意义。

Forecast of Wind Power On-grid Price Based on Wavelet Transforming

A wavelet transforming and neural network forecasting model is presented.First, decompose the history location marginal price sequence into high frequency and low frequency sequences by using wavelet transform,construct the learning samples as the input of a neural network, and predict the sequences of different frequencies with neural networks.Then,achieve wavelet reconstruction and get the theoretical numerical value of LMP.Finally,discuss the results of electricity price on-grid with the analysis of electricity transmission and distribution price.The results show that the forecasting method is effective to forecast the historical location marginal price of the California electricity market,and the analysis of wind power on-grid electricity price mechanism from the angle power market is of important significance.