基于频域分解的短期风电负荷预测

Short-Term Wind Power Load Forecasting Based on Frequency Domain Decomposition

为了克服风电的不规则性和提高风电负荷预测的精度,将频域分解的方法运用在风电负荷预测中可以找到风电的部分规律和在一定程度上克服风电的不规则性。通过对原始负荷数据的频域分解,将数据分解成日周期、周周期、低频和高频四个部分。日周期的部分用神经网络的方法训练和预测。低频部分用一元线性回归的方法预测。高频部分用提升小波和神经网络相结合的方法训练和预测。最后将各部分的预测结果加起来,这样就实现了风电负荷的高精度预测。本文中用实际数据进行仿真,实验结果表明,基于频域分解的方法可以比较好地找到风电的规律,有利于通过不同的方法对不同的部分进行短期负荷预测,很大程度地提高预测的精度,测试表明该方法用于风电负荷预测是有效可行的。

In order to overcome the irregularity of wind power and improve wind power load forecasting accuracy.The frequency domain decomposition methods use in wind power load forecasting can find the wind power law and overcome the irregularity of wind power on a certain extent.With the frequency domain decomposition method,the original load data will be decompose into daily cycle,week cycle,the low and high frequency four parts.Part of the daily cycle will use neural network methods for training and prediction.The low-frequency part will use a linear regression method.The method of combining lifting wavelet and neural network will be used in training and forecasting the high-frequency part.Finally,the various parts of the forecasting results add up to achieve high-precision wind power load forecasting.In this paper,the actual data are used for simulation,the experimental results show that the method based on the frequency domain decomposition,can more easily find the wind power law,benefit short term load forecasting by different methods in different part,greatly improve the precision of forecasting.The test shows that the method used for the wind power load forecast is feasible and effective.