| 基于IPSO-BiLSTM-AM模型的超短期风电功率预测方法 |
| |
| 引用本文： | 高鹭1,孔繁苗1,张飞1,2,任晓颖1,2,张晓琳1,秦岭1. 基于IPSO-BiLSTM-AM模型的超短期风电功率预测方法[J]. 陕西电力, 2022, 0(4): 27-34 |
| |
| 作者姓名： | 高鹭1 孔繁苗1 张飞1 2 任晓颖1 2 张晓琳1 秦岭1 |
| |
| 作者单位： | （1.内蒙古科技大学信息工程学院，内蒙古包头 014010；2.华北电力大学可再生能源学院，北京 102206） |
| |
| 摘 要： | 针对现有模型预测准确性与稳定性较低的问题，提出一种以BiLSTM为基础的风电功率预测模型。BiLSTM可以很好的处理风电多变量之间的非线性关系，其次采用改进的PSO优化BiLSTM的超参数，并通过AM训练模型的权重。最后采用内蒙古自治区某风电场的历史数据进行提前0~15 min试验。结果表明，提出的IPSO-BiLSTM-AM模型具有较高的预测精度，可以为风电场电力调度与控制提供科学参考。
|
| 关 键 词： | 风功率预测 改进的粒子群算法 双向长短期记忆神经网络 注意力机制 |
Ultra Short-term Wind Power Prediction Method Based on IPSO-BiLSTM-AM Model |
| |
| GAO Lu1,KONG Fanmiao1,ZHANG Fei1,2,REN Xiaoying1,2,ZHANG Xiaolin1,QIN Ling1. Ultra Short-term Wind Power Prediction Method Based on IPSO-BiLSTM-AM Model[J]. Shanxi Electric Power, 2022, 0(4): 27-34 |
| |
| Authors: | GAO Lu1 KONG Fanmiao1 ZHANG Fei1 2 REN Xiaoying1 2 ZHANG Xiaolin1 QIN Ling1 |
| |
| Affiliation: | （1. Institute of Information Engineering， Inner Mongolia University of Science and Technology， Baotou 014010， China；2. Renewable Energy College，North China Electric Power University， Beijing 102206， China） |
| |
| Abstract: | In view of the low accuracy and stability of existing models， a wind power prediction model based on BiLSTM is proposed. BiLSTM can handle the non-linear relationship among wind power multi-variables well. Secondly，the improved PSO algorithm is used to optimize the hyper parameters of BiLSTM， and the AM is used for training the weight of the model. Finally， a 0~15 minute advance test is carried out with the historical data of a wind farm in Inner Mongolia Autonomous Region. The results show that the proposed IPSO-BiLSTM-AM model has high prediction accuracy，which can provide scientific reference for the power dispatch and control of wind farms. |
| |
| Keywords: | wind power prediction improved particle swarm optimization bidirectional long short-term memory neural network attention mechanism |
|
| 点击此处可从《陕西电力》浏览原始摘要信息 |
|
点击此处可从《陕西电力》下载免费的PDF全文 |
