Abstract : The Radial Basis Function (RBF) neural network is a feed-forward artificial neural network with strong approximation capability. A K-means algorithm based on density parameter was introduced to determine clustering center aimed to improve the training rate of the RBF. It could reduce sensitivity of traditional K-means algorithm for initial clustering centers. A rainfall forecasting model of RBF based on K-means algorithm was built, which was applied to forecast monthly rainfall in Shuangyashan City during the flood season, aiming to test the effectiveness of this model. The case study showed that the mean relative error of rainfall forecasting in flood season (from June to September) of the year 2006, 2007 and 2008 was 10.81%, and the deterministic coefficient was 0.95. It demonstrated a higher forecasting accuracy comparing to a RBF model based on a standard K-means algorithm and BP (Back Propagation) model, and the rainfall forecasting results satisfied the requirements of hydrologic prediction.
Daoliang Li; Yingyi Chen. 8th International Conference on Computer and Computing Technologies in Agriculture (CCTA), Sep 2014, Beijing, China. IFIP Advances in Information and Communication Technology, AICT-452, pp.218-225, 2015, Computer and Computing Technologies in Agriculture VIII. 〈10.1007/978-3-319-19620-6_27〉
https://hal.inria.fr/hal-01420235
Contributeur : Hal Ifip
<>
Soumis le : mardi 20 décembre 2016 - 14:03:23
Dernière modification le : vendredi 1 décembre 2017 - 01:16:51
Document(s) archivé(s) le : mardi 21 mars 2017 - 13:12:03
Zhenxiang Xing, Hao Guo, Shuhua Dong, Qiang Fu, Jing Li. RBF Neural Network Based on K-means Algorithm with Density Parameter and Its Application to the Rainfall Forecasting. Daoliang Li; Yingyi Chen. 8th International Conference on Computer and Computing Technologies in Agriculture (CCTA), Sep 2014, Beijing, China. IFIP Advances in Information and Communication Technology, AICT-452, pp.218-225, 2015, Computer and Computing Technologies in Agriculture VIII. 〈10.1007/978-3-319-19620-6_27〉. 〈hal-01420235〉
