XGBoost与GRU模型在发电功率预测中的应用  

Application of XGBoost and GRU Model in Power Generation Prediction

作　　者：李光环 杨小天 刘训钊 LI Guanghuan;YANG Xiaotian;LIU Xunzhao(Jilin Architectural University,college of electrical and computer in changchun,Changchun,China,130000;Jilin Normal University,Changchun,China,130000;National energy Changyuan Jingmen power generation Co.,LTD.,Jingmen,China,448000)

机构地区：[1]吉林建筑大学电气与计算机学院,吉林长春130000 [2]吉林师范大学,吉林长春130000 [3]国能长源荆门发电有限公司,湖北荆门4480000

出　　处：《福建电脑》2024年第6期21-26,共6页Journal of Fujian Computer

基　　金：吉林省科技厅项目《大型厂区智能安全管理系统研究与应用示范》(No.20230203139SF)资助。

摘　　要：超短期光伏发电功率预测是电网安全调度与平稳运行的基础。针对传统单一预测模型存在预测精度不高、模型抗干扰能力差、鲁棒性不强等问题,本文提出了一种基于双重XGBoost-GRU的混合预测模型,并经过以澳大利亚爱丽丝泉光伏发电系统为算例的仿真分析。仿真实验的结果表明,双重XGBoost-GRU组合预测模型相较于GRU、XGBoost-GRU误差分别降低了93.93%、75.65%,说明了基于特征创造XGBoost-GRU融合预测模型具有良好的预测效果。The prediction of ultra short term photovoltaic power generation is the foundation for safe dispatch and stable operation of the power grid.In response to the problems of low prediction accuracy,poor anti-interference ability,and weak robustness of traditional single prediction models,this paper proposes a hybrid prediction model based on dual XGBoost GRU,and conducts simulation analysis on the Alice Spring photovoltaic power generation system in Australia.The simulation results show that the double XGBoost GRU combined prediction model reduces errors by 93.93%and 75.65%compared to GRU and XGBoost GRU,respectively,indicating that the feature based XGBoost GRU fusion prediction model has good predictive performance.

关 键 词：GRU模型 XGBoost模型:光伏功率预测 

分 类 号：TM614[电气工程—电力系统及自动化]