基于LSTM的风电功率预测方法研究
Power Prediction Method of Wind Power Based on LSTM
张胜杰 1张雪锋 2谢忠洲 1贾宛英 2张志艳2
作者信息
- 1. 中广核新能源河南有限公司,河南 郑州 450000
- 2. 郑州轻工业大学电气信息工程学院,河南 郑州 450002
- 折叠
摘要
风力发电是中国能源供应体系中的重要组成部分,尤其在风光互补照明系统中发挥着至关重要的作用.为提高风电功率预测的稳定性,构建了基于LSTM(Long Short-Term Memory,长短期记忆网络)的风电功率预测模型.首先,对风电功率数据进行了预处理;然后,建立了基于门控机制的LSTM预测模型,该模型能够捕捉风电功率数据的时序特征和长期依赖关系,实现对风电功率的预测;最后,通过算例对比分析,验证了LSTM预测模型在风电功率预测中具有较好的精度和稳定性.
Abstract
Wind power generation is an important part of China's energy supply system,especially in the wind-solar complementary lighting system plays a vital role.In order to improve the stability of wind power,a wind power prediction model based on LSTM(Long Short-Term Memory)was constructed.Firstly,the wind power data was preprocessed;then,a LSTM prediction model based on gate control mechanism was established,which can capture the time series characteristics and long-term dependence in wind power data,so as to realize the prediction of wind power;finally,an example was given to show that the LSTM prediction model has better accuracy and stability in wind power prediction.
关键词
风电/功率预测/长短期记忆网络Key words
wind power/power prediction/long short-term memory引用本文复制引用
基金项目
河南省科技攻关项目(242102241030)
河南省科技攻关项目(222102320074)
出版年
2024
NETL
NSTL
万方数据