%0 Journal Article
%T 基于灰狼优化算法的LSTM风速预测方法研究<br>Wind Speed Prediction Model Based on GWO-LSTM
%A 谭铮瑜
%J Artificial Intelligence and Robotics Research
%P 532-540
%@ 2326-3423
%D 2024
%I Hans Publishing
%R 10.12677/airr.2024.133055
%X 风力发电的主要决定因素是风速&#65292;而风速具有不稳定性特征&#65292;准确预测风速对调度风力发电系统&#65292;提高风能转换效率等具有非常重要的意义。针对目前风速预测中&#65292;智能模型参数设置难的问题&#65292;本文提出了一种基于灰狼优化算法(GWO)对长短期记忆模型(LSTM)超参数进行寻优的GWO-LSTM模型&#65292;并利用Halifax风速数据集进行风速的预测。实验结果表明&#65292;GWO-LSTM在预测风速上能获得更高的精度&#65292;在均方根误差(RMSE)、平均绝对误差(MAE)及相关系数<i>R</i><sup>2</sup>上&#65292;比传统的LSTM、GRU、Simple RNN效果更优&#65292;这充分说明了本研究提出的GWO-LSTM算法在风速预测上有较好的应用前景。<br>The primary factor of wind energy production is wind speed, which is unstable. For scheduling wind power-producing systems and increasing wind energy conversion efficiency, accurate wind speed forecasting is crucial. This paper proposes a GWO-LSTM model based on the gray wolf optimization algorithm (GWO) to find the optimization of the hyperparameters of the long and short-term memory model (LSTM), and uses the Halifax wind speed dataset for wind speed prediction in order to address the issue that it is difficult to set the parameters of intelligent models in wind speed prediction. The experimental results demonstrate that the GWO-LSTM algorithm proposed in this study has better application prospects in wind speed prediction because it can predict wind speed with a higher degree of accuracy and performs better than the traditional LSTM, GRU, and Simple RNN in terms of root mean square error (RMSE), mean absolute error (MAE), and correlation coefficient (<i>R</i><sup>2</sup>).
%K 灰狼优化算法&#65292
%K 超参数&#65292
%K 长短期记忆网络&#65292
%K 风速&#65292
%K 时间序列预测<br>Gray Wolf Optimization
%K Hyperparameter
%K LSTM
%K Wind Speed
%K Time Series Prediction
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=94152