首页
学术期刊
论文检测
AIGC检测
热点
更多
数据
WIND DIRECTION PREDICTION ALGORITHM BASED ON CEEMDAN AND TIMPORAL CONVOLUTIONAL NETWORK
被引:0
|
作者
:
Zhang, Qun
论文数:
0
引用数:
0
h-index:
0
机构:
State Grid Electric Power Research Institute, Nanjing,211106, China
State Grid Electric Power Research Institute, Nanjing,211106, China
Zhang, Qun
[
1
]
Hou, Yuqiang
论文数:
0
引用数:
0
h-index:
0
机构:
State Grid Electric Power Research Institute, Nanjing,211106, China
State Grid Electric Power Research Institute, Nanjing,211106, China
Hou, Yuqiang
[
1
]
Xu, Jianbing
论文数:
0
引用数:
0
h-index:
0
机构:
State Grid Electric Power Research Institute, Nanjing,211106, China
State Grid Electric Power Research Institute, Nanjing,211106, China
Xu, Jianbing
[
1
]
Zhao, Wei
论文数:
0
引用数:
0
h-index:
0
机构:
State Grid Electric Power Research Institute, Nanjing,211106, China
State Grid Electric Power Research Institute, Nanjing,211106, China
Zhao, Wei
[
1
]
Li, Wei
论文数:
0
引用数:
0
h-index:
0
机构:
State Grid Electric Power Research Institute, Nanjing,211106, China
State Grid Electric Power Research Institute, Nanjing,211106, China
Li, Wei
[
1
]
Liu, Fusuo
论文数:
0
引用数:
0
h-index:
0
机构:
State Grid Electric Power Research Institute, Nanjing,211106, China
State Grid Electric Power Research Institute, Nanjing,211106, China
Liu, Fusuo
[
1
]
机构
:
[1]
State Grid Electric Power Research Institute, Nanjing,211106, China
来源
:
Taiyangneng Xuebao/Acta Energiae Solaris Sinica
|
2024年
/ 45卷
/ 10期
关键词
:
Convolution - Deep learning - Prediction models - Random forests - Wind forecasting;
D O I
:
10.19912/j.0254-0096.tynxb.2023-0928
中图分类号
:
学科分类号
:
摘要
:
In order to improve the accuracy of wind direction prediction,a combined prediction algorithm based on decision tree method (CART),random forest algorithm,complete adaptive noise empirical mode decomposition (CEEMDAN) and temporal convolutional network(TCN)is proposed. Among them,the input importance evaluation based on decision tree method is used to evaluate and screen the input relevance of wind direction perdiction models. Randon forest algorithm is used to classify and process wind direction data;The complete adaptive noise integrated empirical mode decomposition is used to decompose the input wind direction data and extract the input information features;Finally,the temporal convolutional network is used to build the wind direction prediction model. The experimental results show that compared to the other eight comparison models,the prediction errors of the proposed model for wind direction in the fourth quarter data set are less than 4.95°,and the highest prediction accuracy is achieved. © 2024 Science Press. All rights reserved.
引用
收藏
页码:512 / 520
相关论文
未找到相关数据