Open AccessPrediction Method of PV Output Power Based on Cloud Model
Author(s) -
Chen Zhong,
Che Songyang,
Xu Yan,
Yin Dapeng
Publication year - 2017
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2017.0585
Subject(s) - cloud computing , computer science , generator (circuit theory) , sample (material) , power (physics) , data mining , physics , quantum mechanics , operating system , chemistry , chromatography
The output of photo‐voltaic power generation is influenced by many factors, and the impacts of different factors are not the same. To establish a precise mathematical model and a stable prediction analysis is safe for power grid. This study, according to the factors of environmental temperature, environmental humidity and radiation intensity, proposed a PV output forecast model based on the cloud model in the short term, which defined a month as a cycle and an hour as a unit. Firstly, the cloud‐transformation model was established to transform factors into some concepts, and the concepts were combined to less by cloud‐combination model, then uncertain factors were achieved. Furthermore, the relationship between each factor was found. At last, the rule generator was established to predict the output of photo‐voltaic power generation. In this paper, we establish cloud model to expand the sample size and increase the number of cloud concepts to improve the accuracy of results effectively. It shows that the relative error is reduced compared with that of other prediction models, so cloud model will be of great value to prediction.