摘要
由于风速具有随机性和间歇性的特点,以传统方法难以实现风速的精准测量及预测。风速信号对于风电机组输出功率稳定、电能质量提升优化等具有重要作用。基于此,提出一种基于卷积与循环神经网络相结合的深度学习实现风速预测的方法,并与其它方法做了对比分析。以某风电场2014—2015年机组历史大数据为依据,经过数据预处理随机选取44天数据对设计模型进行训练验证,结果与实际风速基本一致,并且效果好于其他方法。从该风场2015年历史大数据中再随机选取12天数据,进一步对模型泛化性能进行检验,结果表明该模型仍然能够实现风速的准确预测,泛化性能良好。
Since the wind speed is stochastic and intermittent,it is difficult to measure or predict wind speed with high accuracy by traditional methods.Wind speed signal plays an important role in stabling output power,improving and optimizing the power quality of wind turbines.To solve the problem,a wind speed prediction method based on deep learning combined by convolutional neural network and recurrent neural network was proposed.Based on the big data of a certain wind farm from 2014 to 2015,the proposed model was trained and verified by 44-day data selected randomly from the processed data.The experimental results show that the wind speed prediction results of the proposed model are basically consistent with the actual wind speed.Compared with other prediction methods,the proposed method has the best results.The generalization performance of the proposed model was further verified by 12-day data selected randomly from the big data of the wind farm in 2015.The results show that the model can still predict the wind speed accurately and has excellent generalization performance.
作者
李大中
李颖宇
王超
LI Dazhong;LI Yingyu;WANG Chao(School of Control and Computer Engineering,North China Electric Power University,Baoding 071003,China)
出处
《电力科学与工程》
2019年第8期1-6,共6页
Electric Power Science and Engineering
基金
河北省自然科学基金资助项目(F20170629-23)
关键词
风电机组
风速预测
卷积循环神经网络
深度学习
wind turbine
wind speed prediction
convolutional recurrent neural network
deep learning
