摘要
为了提高光伏发电预测的准确性,优化光伏发电,达到对能源的充分合理利用,光伏数据必须有良好的质量,数据清洗尤为重要。文中提出了一种基于三次样条插值和皮尔逊相关的光伏数据清洗方法。首先删除冗余数据并对异常数据进行判定,再根据光伏数据的特性,针对不同异常数据进行结合三次样条插值和皮尔逊相关的数据重构。Matlab仿真结果表明本清洗方法能有效过滤异常并实现对异常数据的重构,与其他常用清洗方法相比本清洗方法的数据利用率和重构正确率更高。
In order to improve the accuracy of photovoltaic power generation prediction,optimize photovoltaic power generation,and achieve full and reasonable utilization of energy,photovoltaic data must have good quality,data cleaning is particularly important. This paper proposes a photovoltaic data cleaning method based on cubic spline interpolation and Pearson correlation. Firstly,the redundant data is deleted and the abnormal data is determined. Then,based on the characteristics of the photovoltaic data,cubic spline interpolation and Pearson-related data cleaning are performed for different abnormal data. The Matlab simulation results show that the cleaning method can effectively filter the anomaly and reconstruct the abnormal data. Compared with other common cleaning methods,the data utilization and reconstruction accuracy of this method is higher.
作者
肖心园
江冰
任其文
尹晓东
卢静
XIAO Xin-yuan;JIANG Bing;REN Qi-wen;YIN Xiao-dong;LU Jing(College of Computer Internet of Things Engineering,Hohai University,Changzhou 213022,Jiangsu Province,China;Shandong Electric Power Engineering Consulting Institute Corp., Ltd. (SDEPCI),Jinan 250013,China)
出处
《信息技术》
2019年第5期19-22,28,共5页
Information Technology
基金
江苏省产业前瞻与共性关键技术科技项目(BE2017063)
关键词
数据清洗
光伏数据
三次样条插值
皮尔逊相关
data cleaning
photovoltaic data
cubic spline interpolation
pearson correlation
