科研之友
微信
新浪微博
Facebook
分享链接
摘要
The ecology is the science of study on structure and function of ecosystem, within a certain range of biological species, quantity, biomass, life history and spatial distribution; environmental factors on biological and biological reaction to the environment; and how many certain relationship between organisms living environment moisture level and precipitation. Among them, the forecast of precipitation is one of the most important and challenging task. In general, precipitation is highly non-linear and complicated phenomena, which require advanced computer modelling and simulation for the accurate prediction. A new hybrid model is proposed based on Ensemble empirical mode decomposition (EEMD) and General regression neural network (GRNN). Firstly, precipitation is broken down into series to different scales intrinsic mode function; secondly, put these components as the input of the GRNN; finally, the final prediction value is predicted by the new algorithm. Next, the model is tested on annual precipitation sequence from 2007 to 2011 in Zhengzhou, and compares the forecast results obtaining from EMD-GRNN model and the EEMD-GRNN model. The result shows that the latter predictive value is close to the actual precipitation, which can better reflect the actual precipitation change.
- 出版日期2016
- 单位西安理工大学
