基于长短期记忆神经网络的风速超短期快速滚动预报技术  

Research on Ultra Short-Term Fast Rolling Prediction Technology of Wind Speed Based on LSTM Neural Network

作　　者：方楠 姜舒婕 闫晓敏 阮小建 马辛宇[4] FANG Nan;JIANG Shujie;YAN Xiaomin;RUAN Xiaojian;MA Xinyu(Early Warning Center in Zhejiang,Hangzhou 310052;Weather Modification Center in Zhejiang,Hangzhou 310052;Gansu Provincial Meteorological Service Center,Lanzhou 730020;Zhejiang Provincial Meteorological Service Center,Hangzhou 310052)

机构地区：[1]浙江省预警信息发布中心,杭州310052 [2]浙江省人工影响天气中心,杭州310052 [3]甘肃省气象服务中心,兰州730020 [4]浙江省气象服务中心,杭州310052

出　　处：《气象科技》2022年第6期842-850,共9页Meteorological Science and Technology

基　　金：浙江省气象科技青年项目(2021QN07);甘肃省气象局科研项目(Ms2022-22)共同资助。

摘　　要：利用甘肃省某风电场2017—2020年测风数据,基于长短期记忆神经网络(LSTM)模型,通过评估不同输入数据和模型时间窗口长度下的预报精度,设计一套适用于风电场的风速超短期快速滚动预报方案。结果表明:通过输入不同的特征变量,在风速的超短期(未来4 h内)预报中,风速自身变化起主导作用,模型输入变量中只加入各高度层的风速能得到更好的模拟效果。通过评估LSTM模拟时间窗口长度L对模拟效果的影响,当时间窗口长度L≤24 h时,模拟效果较好,说明超短期风速变化主要和风速自身临近时刻的变化有关;当L>24 h时,模拟效果快速下降,说明过长的L会削弱模拟能力,降低模拟精度。通过分析LSTM在未来4 h内的风速模拟能力,发现随着预报时长的增加,模拟精度逐步下降,但在未来2 h内的风速均方根误差RMSE均小于2 m·s^(-1),结果较为理想,且该方法对计算资源要求不高,经济实用性强,在业务中具有较高的应用潜力。Using the observation data of a certain wind farm in Gansu,an ultra short-term fast rolling wind speed forecast method is proposed based on the Long Short-Term Memory(LSTM)neural network model by evaluating the forecast accuracy under different input variables and model time window lengths.The results show that the change in wind speed itself plays a leading role in the ultra short-term wind speed forecast.Better simulation results can be obtained when input variables only include the wind speed data at different altitudes.By evaluating the impact of time window length L of LSTM on simulation capability,it is found that when L≤24 h,the model works well,which means that the change of ultra short-term wind speed is mainly related to the change of its own near time.When L>24 h,the simulation effect of all schemes decreases,which means overly long L reduces the simulation accuracy.By evaluating the wind speed forecast capability of LSTM in the next 4 hours,it is found that the simulation accuracy decreases gradually while the prediction time increases.The forecast ability is ideal in the next 2 hours,and the RMSE is less than 2 m·s^(-1).LSTM proves economical and practical with low requirements for computing resources and has high application potential in operational wind speed forecast practice.

关 键 词：风速预测 长短期记忆神经网络 预报模型 风力发电 时间序列 

分 类 号：P457.5[天文地球—大气科学及气象学]