摘要
针对相似日对光伏功率预测精度的影响,提出基于相似日的Grey-Markov与BP_Adaboost的光伏功率预测方法。为获取不同相似日,分别以辐照度和温度为相似变量,通过二维欧氏距离选取两组相似日;基于两组相似日数据,用灰色GM(1,1)模型预测光伏功率的总体趋势,用马尔科夫链对灰色模型的预测结果进行修正,得到两组预测结果;用BP_Adaboost对两组预测结果进行集成,以获得更高的预测精度。仿真结果表明,该方法提高了结果的预测精度与鲁棒性,可为光伏电站并网提供重要参考信息。
For the influence of similar days on PV power prediction accuracy,the PV power prediction based on Grey-Markov and BP_Adaboost by similar days was proposed.In order to get different similar days,taking irradiance and temperature as similar variables,the two-dimensional Euclidean distance was used to select two groups of similar days.Based on two groups of similar days,the general trend of photovoltaic output power was predicted by the grey GM(1,1)model,and the results predicted by grey model were modified by Markov chain to obtain two groups of forecast results.The two groups of prediction results were integrated with BP_Adaboost to get better prediction results.The simulation results show that the method improves the prediction accuracy and robustness of the results,and it can provide important reference information for the grid connection of photovoltaic power stations.
作者
杨锡运
王诗晨
张艳峰
彭琰
马骏超
YANG Xiyun;WANG Shichen;ZHANG Yanfeng;PENG Yan;MAJunchao(Department of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China;Electric Power Research Institute,State Grid Zhejiang Electric Power Supply Co.,Ltd.,Hangzhou Zhejiang 310014,China)
出处
《电源技术》
CAS
北大核心
2023年第6期790-794,共5页
Chinese Journal of Power Sources
基金
国网浙江省电力有限公司科技项目(5211DS220009)。