基于GWO-CNN-BiLSTM的超短期风电预测  

Ultra-short-term Wind Power Prediction Based on GWO-CNN-BiLSTM

作　　者：程杰[1,2] 陈鼎 李春 钟伟东 严婷 窦春霞 CHENG Jie;CHEN Ding;LI Chun;ZHONG Wei-dong;YAN Ting;DOU Chun-xia(Institute of Advanced Technology for Carbon Neutrality,Nanjing University of Posts and Telecommunications,Nanjing 210023,China;College of Automation&Artificial Intelligence,Nanjing University of Posts and Telecommunications,Nanjing 210023,China;Jiaxing Power Supply Company of State Grid Zhejiang Electric Power Company,Jiaxing 314000,China)

机构地区：[1]南京邮电大学,碳中和先进技术研究院,南京210023 [2]南京邮电大学自动化学院、人工智能学院,南京210023 [3]国网浙江省电力有限公司嘉兴供电公司,嘉兴314000

出　　处：《科学技术与工程》2023年第35期15091-15099,共9页Science Technology and Engineering

基　　金：国网公司总部科技项目(5400-202219152A-1-1-ZN)。

摘　　要：在未来高渗透率风电场景下,超短期风电功率预测研究对于实现电力系统优化运行具有重要意义。为此,提出一种基于GWO-CNN-BiLSTM的超短期风电预测方法。首先,搭建基于卷积神经网络(convolutional neural network,CNN)与双向长短期记忆神经网络(bidirectional long short term memory,BiLSTM)的组合模型,然后,为提升风电预测结果的精度,通过灰狼优化算法(grey wolf optimizer,GWO)对组合模型进行优化,使该组合模型参数能实时适应风电历史数据。最后,仿真结果验证了所提出方法的有效性和优越性。In the future high-permeability wind power scenarios,the study of ultra-short-term wind power prediction research is of great significance for achieving optimal operation of power systems.Therefore,an ultra-short-term wind power prediction method based on GWO-CNN-BiLSTM was proposed.Firstly,a combined model based on convolutional neural network(CNN)and bidirectional long short term memory(BiLSTM)was built,and then,in order to improve the accuracy of wind power prediction results,the combined model was optimized by the grey wolf optimizer(GWO),so that the parameters of the combined model can be adapted to the historical wind power data in real time.Finally,the simulation results verify the effectiveness and superiority of the proposed method.

关 键 词：风电预测 CNN BiLSTM GWO 组合模型 

分 类 号：TM614[电气工程—电力系统及自动化]