摘要
准确的风电预测可以提高电网的稳定性和可靠性,优化风电发电计划,降低能源成本。为了提高短期电力负荷预测的精度,文章探讨了一种基于QPSO算法对LSTM神经网络进行优化的算法,并根据LSTM神经网络以及QPSO算法的基本原理,利用QPSO算法对LSTM的超参数及网络拓扑结构进行优化,建立QPSO-LSTM短期风电负荷预测模型。仿真结果表明,QPSO-LSTM模型较传统的LSTM模型预测精度更高,且具有更快的收敛速度。
Accurate wind power forecasting can enhance the stability and reliability of the power grid,optimize wind power generation plans,and reduce energy costs.In order to bolster the accuracy of short-term power load forecasting,the article aims to optimize LSTM neural network through QPSO algorithm.According to the basic principles of LSTM neural network and QPSO algorithm,a forecasting model is established by optimizing the hyperparameters and network topology of LSTM.The simulation results suggest that the QPSO-LSTM model has higher accuracy and faster convergence speed compared to traditional LSTM models.
作者
谭才兴
岳雨霏
汤赐
Tan Caixing;Yue Yufei;Tang Ci(School of Electrical and Information Engineering,Changsha University of Science&Technology,Changsha 410114)
出处
《中阿科技论坛（中英文）》
2023年第12期88-91,共4页
China-Arab States Science and Technology Forum
基金
长沙理工大学大学生创新创业训练计划项目“高比例风电电力系统储能运行及配置分析”(2023047)
长沙理工大学本科教育“金课”建设项目(线下“金课”)“电机学A”。
