摘要
随着国家碳达峰、碳中和及构建以新能源为主体的新型电力系统任务的提出,未来新能源是电力系统的主力军。对于含大规模风电接入的电力系统,风速及风电功率的准确预测对保证系统安全稳定运行、降低风电消纳成本有着至关重要的作用。为此,基于统计方法和神经网络方法,通过设计一个多神经网络的数据融合算法来预测下一年的每小时风速及发电量。利用张北、内蒙地区两风电场的数据样本对数据集进行训练和测试,结果显示平均绝对误差(MAE)很小,预测效果较好。
With the peak carbon dioxide emissions,carbon neutral and the task of building a new power system with new energy as the mainstay,new energy is the main force of the power system in the future.For the power system containing large-scale wind power access,the accurate prediction of wind speed and wind power plays a crucial role in ensuring the safe and stable operation of the system and reducing the cost of wind power consumption.To this end,a multi-neural network data fusion algorithm is designed to predict hourly wind speed and power generation for the following year based on statistical methods and neural network methods.The data set is trained and tested using data samples from wind farms in the Zhangbei and Neimeng area,and the results show that the mean absolute error(MAE)is small and the prediction effect is good.
作者
李建林
张海军
庞俊强
LI Jian-lin;ZHANG Hai-jun;PANG Jun-qiang(Guohua(Hebei)New Energy Co.,Ltd.,Zhangjiakou 075000)
出处
《环境技术》
2021年第5期219-226,共8页
Environmental Technology
基金
神华集团公司科技项目“新能源中长期发电量预测对电力市场交易的辅助决策研究”,项目编号:SHXNY20190338。
关键词
长期预测
神经网络
风能
风速
功率
long-term forecasting
neural network
wind energy
wind speed
power
