一种基于决策融合策略的全天空地基云图云量估计方法  

CLOUD COVER ESTIMATION METHOD OF ALL-SKY GROUND-BASED CLOUD IMAGE BASED ON DECISION FUSION STRATEGY

在线阅读下载全文

作  者:方明[1,2,3] 张利箭 Fang Ming;Zhang Lijian(School of Artificial Intelligence,Changchun University of Science and Technology,Changchun 130022,China;School of Computer Science and Technology,Changchun University of Science and Technology,Changchun 130022,China;Zhongshan Institute of Changchun University of Science and Technology,Zhongshan 528403,China)

机构地区:[1]长春理工大学人工智能学院,长春130022 [2]长春理工大学计算机科学技术学院,长春130022 [3]长春理工大学中山研究院,中山528403

出  处:《太阳能学报》2023年第10期245-254,共10页Acta Energiae Solaris Sinica

基  金:国家自然科学基金(U2141231)。

摘  要:云量对光伏发电有至关重要的影响。全天空地基云图云量估计的重点在于对全天空图像进行畸变矫正,以及设计准确率高的地基云分割方法。针对这两个问题,提出一种基于决策融合策略的云量估计模型。首先,通过基于多维球切面透视投影模型的鱼眼畸变矫正算法,获取多方位的局部复原图像;其次,采用最近邻复原中心的决策融合方法对多维复原图像分割后的数据进行修正,最终获取云量估计值。实验表明:该方法更好地复原了云和天空的比例关系,提高了云量估计的准确性。Cloud coverage has a crucial impact on photovoltaic power generation.The key point of cloud cover estimation is to correct the distortion of the all-sky image and to design the segmentation method of ground-based cloud with high accuracy.Aiming at these two problems,this paper proposes a cloud cover estimation model based on decision fusion strategy.Firstly,the fisheye distortion correction algorithm based on the perspective projection model of multi-dimensional spherical section is used to obtain multi-directional local restoration images.Secondly,the decision fusion method of the nearest neighbor restoration center is used to modify the data after multi-dimensional restoration image segmentation,and finally,the estimated cloud cover is obtained.Experimental results show that this method can restore the proportion relationship between cloud and sky better and improve the accuracy of cloud cover estimation.

关 键 词:云量估计 太阳能 畸变校正 球面投影 决策融合 

分 类 号:TP391[自动化与计算机技术—计算机应用技术] TN911.73[自动化与计算机技术—计算机科学与技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象