摘要
为了解决现有负荷预测方法精度不高、时效性较差的问题,设计一种基于双层XGBoost算法的发电机组负荷优化预测方法。采用双层XGBoost模型对发电机组负荷进行精确预测和优化。在第一层模型中,组建弱学习器;在第二层模型中,将第一层模型的输出结果作为输入,细化目标函数,构建双层串行集成学习下的负荷预测模型。使用萤火虫算法对预测模型进行优化解算,以实现发电机组负荷优化预测。经过实验验证,所提出的方法具有较低的平均绝对误差和均方根误差,并且预测时间控制在可接受范围内。
In order to solve the problems of low accuracy and poor timeliness of existing load prediction methods,a load optimization prediction method for generator set is designed based on double layer XGBoost algorithm.Double layer XGBoost model is used to accurately predict and optimize the load of generator set.In the first layer model,a weak learner is constructed.In the second layer model,the output results of the first layer model are used as inputs,and the objective function is refined to construct a load prediction model under double layer serial ensemble learning.The prediction model is optimized by using the firefly algorithm to achieve optimal load prediction for generator set.After experimental verification,the proposed method has low average absolute error and root mean square error,and the prediction time is controlled within an acceptable range.
作者
罗潘
何心铭
田闽哲
付航
申志刚
LUO Pan;HE Xinming;TIAN Minzhe;FU Hang;SHEN Zhigang(State Grid Henan Electric Power Company,Zhengzhou 450000,China;State Grid Henan Electric Power Company Marketing Service Center,Zhengzhou 450000,China;Henan Jiuyu Tenglong Information Engineering Co.,Ltd.,Zhengzhou 450000,China)
出处
《微型电脑应用》
2025年第12期54-57,74,共5页
Microcomputer Applications
基金
国网河南省电力公司科技项目(52170217001J)。
