引入SMA - IQR和注意力机制的TCN - GRU模型超短期风电功率预测
郑哲明 , 孔玲玲 , 何印
云南民族大学学报(自然科学版) ›› 2025, Vol. 34 ›› Issue (04) : 480 -486.
引入SMA - IQR和注意力机制的TCN - GRU模型超短期风电功率预测
TCN-GRU model with SMA-IQR and attention mechanism for ultra-short term wind power prediction
随着风力发电机大规模接入电网,风速的间歇性、波动性等特点导致风电功率的输出极不稳定.针对超短期风电功率预测精度的问题提出了一种基于简单移动平均法数据去噪(simple moving average, SMA)和四分位法(interquartile range, IQR)清洗检测异常数据的时间卷积网络(temporal convolutional network,TCN)、门控循环单元(gated recurrent unit, GRU)和注意力机制(attention mechanism,AM)超短期风电功率预测模型.实验结果表明,模型的平均绝对值误差、平均绝对值百分比误差、决定系数比其他单一模型和组合模型具有更高的精确度,验证了该模型具备更高效的超短期风电功率预测能力.
As wind turbines are connected to the power grid on a large scale, the intermittently fluctuating wind speed and other characteristics lead to extremely unstable wind power output. To solve the problem of ultra - short - term wind power prediction accuracy, a temporal convolutional network based on simple moving average data denoising (SMA) and interquartile range (IQR) cleaning and detecting abnormal data is proposed. TCN, gated recurrent unit (GRU) and attention mechanism (AM) for ultra - short term wind power prediction models. Experimental results show that, compared with other prediction models, the mean absolute error (MAE) and mean absolute percentage error (MAPE) of the model in this paper are as follows: MAPE and R - Square have higher accuracy than other single models and combined models, verifying that the model has a more efficient ultra - short term wind power prediction ability.
时间卷积网络 / 门控循环单元 / 风电功率 / 注意力机制 / 四分位法 / 移动平均
time convolutional network / gated cycle unit / wind power / attention mechanism / quartile method / moving average
| [1] |
唐西胜,苗福丰,齐智平, |
| [2] |
张淑清,杨振宁,姜安琦, |
| [3] |
洪翠,林维明,温步瀛.风电场风速及风电功率预测方法研究综述[J].电网与清洁能源,2011,27(1):60 - 66. |
| [4] |
|
| [5] |
郑婷婷,王海霞,李卫东.风电预测技术及其性能评价综述[J].南方电网技术,2013,7(2):104 - 109. |
| [6] |
杨海民,潘志松,白玮.时间序列预测方法综述[J].计算机科学,2019,46(1):21 - 28. |
| [7] |
|
| [8] |
杨茂,周宜.计及风电场状态的风电功率超短期预测[J].中国电机工程学报,2019,39(5):1259 - 1268. |
| [9] |
|
| [10] |
陈简,张惊雷.基于注意力机制的TCN - BiGRU模型短期风电功率预测[J].天津理工大学学报,2024,3(29):1673 - 095X. |
| [11] |
郭玲,徐青山,郑乐.基于TCN - GRU模型的短期负荷预测方法[J].电力工程技术,2021,40(3):66 - 71. |
| [12] |
王凇瑶,张智晟.基于量子加权多层级GRU神经网络的综合能源系统多元负荷短期预测[J].电力系统保护与控制,2022,50(23):85 - 93. |
| [13] |
张健,田海.基于DE - XGBoost的短期风电功率预测[J].信息技术,2024,7(22):1009 - 2552. |
| [14] |
李相俊,许格健.基于长短期记忆神经网络的风力发电功率预测方法[J].发电技术,2019,40(5):426 - 433. |
| [15] |
|
| [16] |
|
| [17] |
|
| [18] |
|
| [19] |
|
| [20] |
|
| [21] |
|
国家自然科学基金(52061042)
云南民族大学硕士研究生科研基金(2025SKY025)
2024年度云南省专业学位研究生教学案例库建设项目《电气类专业工程伦理案例库建设》
/
| 〈 |
|
〉 |