基于改进XGBoost算法的配网线路光伏发电功率预测方法
Power Prediction Method of Photovoltaic Power Generation on Distribution Lines Based on Improved XGBoost Algorithm
赵铭滔1
作者信息
- 1. 国网兰溪市供电公司,浙江金华 321000
- 折叠
摘要
针对现有预测方法在对配网线路光伏发电功率进行预测时存在预测结果与实际情况相差较大,不符合预测精度要求的问题,引入改进XGBoost算法,开展对配网线路光伏发电功率预测方法的设计研究.分析影响配网线路光伏发电功率的因素,选择变量特征;基于改进XGBoost算法,对各个变量特征的贡献度计算,并利用预测模型完成对发电功率的预测.实验结果表明:新的预测方法得到的预测结果与实际情况更相符,具备极高的预测精度.
Abstract
Aiming at the existing prediction methods in the prediction of photovoltaic power of distribution lines,there are problems that the prediction results differ greatly from the actual situation and do not meet the requirements of prediction accuracy,the improved XGBoost algorithm is introduced to carry out the research on the design of the prediction method of photovoltaic power of distribution lines.Analyze the factors affecting the PV power generation of distribution lines and select variable features;based on the improved XGBoost algorithm,the contribution degree of each variable feature is calculated,and the prediction model is used to complete the prediction of power generation.The experimental results show that the prediction results obtained by the new prediction method are more consistent with the actual situation and have very high prediction accuracy.
关键词
改进XGBoost算法/光伏/预测/发电功率/配网线路Key words
improved XGBoost algorithm/photovoltaic/prediction/generation power/distribution line引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据