基于小波分析和改进BP神经网络的风电功率预测算法

A Wind Power Prediction Algorithm Based on Wavelet Analysis and Improved BP Neural Network

由于风力间歇性、随机性以及不可控性等特点,风电的输出功率不稳定,波动范围较大,风电的这些特点势必会对电网的供电质量造成不利影响。对风电的输出功率进行预测有利于发电场站及时调整发电计划,从而提高供电质量。本文提出一种基于小波分析和改进BP神经网络的风电功率预测算法,利用小波分析算法对风速时间序列进行预处理,再输入BP神经网络进行预测,得到风速预测值后根据风电机组的功率特性曲线计算风电功率的预测值。与单纯BP神经网络模型相比具有更高的预测精度。实验表明,本文所提算法具有优良的预测精度,具备部署到风电场站实际运用的可行性。

Due to the intermittent,random and uncontrollable characteristics of wind power,the output power of wind farms is unstable and exhibits considerable fluctuation.These characteristics of wind power are bound to adversely affect the power supply quality of power grid.Prediction of the wind power output makes for timely adjustment of power generation plans of power generation stations,so as to improve the quality of power supply.In this paper,a wind power prediction algorithm based on wavelet analysis and improved BP neural network is proposed.The wavelet analysis algorithm is used to preprocess the wind speed time series which is then input into the BP neural network for prediction.After the predicted wind speed is obtained,the predicted wind power is calculated according to the power characteristic curve of the wind turbine.Compared with the simple BP neural network model,the proposed algorithm has higher prediction accuracy.The experimental results show that the proposed algorithm has excellent prediction accuracy and is feasible to be deployed in wind farms.