Study On The Precipitation Forecast Based On Neural Network And Wavelet Analysis

Jiangsu Province is located in the Yangtze-Huaihe River basin, which is one of the areas most severely affected by droughts and floods. Therefore, it is important to study on the precipitation forecast in this area for the flood control with drought relief.In this paper, the precipitation forecast models based on back propagation neural network, radial basis function neural network and wavelet neural network are established with the precipitation data from four rainfall stations in Jiangsu Province, which include the Xuzhou Station during the year of1960-2009, the Ganyu Station during the year of1957-2009, the Dongtai Station during the year of1953-2009and the Nanjing Station during the year of1951-2009.The following conclusions can be obtained by the example analysis:(1) The annual precipitation series of Xuzhou station is the dependent sequence. The wavelet neural network precipitation forecast model works best, the radial basis function neural network precipitation forecast model ranks secondly, the back propagation neural network precipitation forecast model is the poorest.(2) The annual precipitation series of Ganyu Station, Dongtai Station and Nanjing Station are the independent sequences. And the forecast results of the back propagation neural network precipitation forecast model are more precise than the wavelet neural network precipitation forecast model in these three stations.(3) It is feasible to apply the back propagation neural network precipitation forecast model, the radical basis function neural network precipitation forecast model and the wavelet neural network precipitation forecast model. To a certain extent, these models can reflect the trends of the precipitation.(4) These models can be improved according to their demerits and coupled combined with their merits in the near future. In order to improve the precision of precipitation forecast, the mutual verifying and checking of these models should be considered in practice.