摘要
在收集吕梁市光伏电站发电历史数据和历史气象数据的基础上,首先对历史数据进行整合,建立历史数据样本集;其次对整合好的数据进行数据预处理,选择出影响光伏发电的主要因素,并进行赋值;再次通过等级比较算法筛选出与预测日条件因素相同或高度相似的历史数据序列;最后将预测日气象等级与历史数据库对比得出预测日光伏电站的发电功率。预测结果表明,2021年1月至2021年4月每个月的发电量预测误差分别为7.7%,9.5%,8.3%,5.4%,4个月的平均预测误差为7.7%。
Based on the collection of historical power generation data and historical meteorological data of the photovoltaic power station in Lvliang City,this paper firstly integrated the historical data to establish the historical data sample set;secondly,it preprocessed the integrated data to select the main factors affecting photovoltaic power generation and assigned values to them;again,it screened the historical data series with the same or highly similar condition factors as the forecast day by the rank comparison algorithm;finally,it derived the power generation of the photovoltaic power station on the forecast day by comparing the meteorological rating of the forecast day with the historical database.The forecast results showed that the forecast errors of power generation for each month from January 2021 to April 2021 were 7.7%,9.5%,8.3%and 5.4%respectively,and the average forecast error for the four months was 7.7%.
作者
杨洁
韩郁
舒鹏丽
YANG Jie;HAN Yu;SHU Pengli(Department of Physics,Lvliang University,Lvliang 033000,Shanxi,China)
出处
《能源与节能》
2022年第12期7-9,18,共4页
Energy and Energy Conservation
基金
吕梁市重点研发计划(高新技术领域)项目(2020GXZ DYF21)。
关键词
光伏电站
发电功率
预测
数据预处理
photovoltaic power station
power generation capacity
prediction
data preprocessing
作者简介
第一作者:杨洁,1989年生,女,山西吕梁人,2015年毕业于西北工业大学热能工程专业,硕士,讲师。
