Modified Particle Swarm Optimization with Attention-Based LSTM for Wind Power Prediction

被引:31
作者
Sun, Yiyang [1 ]
Wang, Xiangwen [2 ]
Yang, Junjie [3 ]
机构
[1] Shanghai Univ Elect Power, Coll Elect & Informat Engn, 185,Hucheng Ring Rd, Shanghai 201306, Peoples R China
[2] Shanghai Univ Elect Power, Coll Elect & Informat Engn, 2103 Pingliang Rd, Shanghai 200090, Peoples R China
[3] Shanghai Dianji Univ, Sch Elect Informat Engn, 300 Shuihua Rd, Shanghai 201306, Peoples R China
基金
中国国家自然科学基金;
关键词
wind power prediction; modify particle swarm optimization algorithm (MPSO); attention mechanism; LSTM neural network; MODEL;
D O I
10.3390/en15124334
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
The accuracy of wind power prediction is crucial for the economic operation of a wind power dispatching management system. Wind power generation is closely related to the meteorological conditions around wind plants; a small variation in wind speed could lead to a large fluctuation in the extracted power and is difficult to predict accurately, causing difficulties in grid connection and generating large economic losses. In this study, a wind power prediction model based on a long short-term memory network with a two-stage attention mechanism is established. An attention mechanism is used to measure the input data characteristics and trend characteristics of the wind power and reduce the initial data preparation process. The model effectively alleviates the intermittence and fluctuation of meteorological conditions and improves prediction accuracy significantly. In addition, the modified particle swarm optimization algorithm is introduced to optimize the hyperparameters of the LSTM network, which speeds up the convergence of the model dramatically and avoids falling into local optima, reducing the influence of man-made random selection of LSTM network hyperparameters on the prediction results. The simulation results on the real wind power data show that the modified model has increased prediction accuracy compared with the previous machine learning methods. The monitoring and data collecting system for wind farms reveals that the accuracy of the model is around 95.82%.
引用
收藏
页数:17
相关论文
共 31 条
[1]  
Bahdanau D, 2016, Arxiv, DOI arXiv:1409.0473
[2]   LEARNING LONG-TERM DEPENDENCIES WITH GRADIENT DESCENT IS DIFFICULT [J].
BENGIO, Y ;
SIMARD, P ;
FRASCONI, P .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (02) :157-166
[3]   Multi-Sequence LSTM-RNN Deep Learning and Metaheuristics for Electric Load Forecasting [J].
Bouktif, Salah ;
Fiaz, Ali ;
Ouni, Ali ;
Serhani, Mohamed Adel .
ENERGIES, 2020, 13 (02)
[4]   Single and Multi-Sequence Deep Learning Models for Short and Medium Term Electric Load Forecasting [J].
Bouktif, Salah ;
Fiaz, Ali ;
Ouni, Ali ;
Serhani, Mohamed Adel .
ENERGIES, 2019, 12 (01)
[5]   A sensitivity study of the WRF model in wind simulation for an area of high wind energy [J].
Carvalho, David ;
Rocha, Alfredo ;
Gomez-Gesteira, Moncho ;
Santos, Carlos .
ENVIRONMENTAL MODELLING & SOFTWARE, 2012, 33 :23-34
[6]  
Cho K., 2014, PROC 8 WORKSHOP SYNT, P103, DOI DOI 10.3115/V1/W14-4012
[7]   Models for Short-Term Wind Power Forecasting Based on Improved Artificial Neural Network Using Particle Swarm Optimization and Genetic Algorithms [J].
Dinh Thanh Viet ;
Vo Van Phuong ;
Minh Quan Duong ;
Quoc Tuan Tran .
ENERGIES, 2020, 13 (11)
[8]  
Viet DT, 2019, INT CONF SYST SCI EN, P190, DOI [10.1109/ICSSE.2019.8823479, 10.1109/icsse.2019.8823479]
[9]   Numerical Weather Prediction and Artificial Neural Network Coupling for Wind Energy Forecast [J].
Donadio, Lorenzo ;
Fang, Jiannong ;
Porte-Agel, Fernando .
ENERGIES, 2021, 14 (02)
[10]   The Wind Integration National Dataset (WIND) Toolkit [J].
Draxl, Caroline ;
Clifton, Andrew ;
Hodge, Bri-Mathias ;
McCaa, Jim .
APPLIED ENERGY, 2015, 151 :355-366