Precipitation prediction in shenzhen city based on WNN

Rainfall is a key factor in development of a city. Accurate prediction of a city's rainfall can prevent the occurrence of waterlogging. In traditional prediction method, most experts are from the point of view of statistics and time series. However, due to the nonstationarity and non-linearity of rainfall, the prediction result is not ideal. Based on traditional BP neural network and the multiscale and nonstationarity of wavelet analysis, the nonlinearity of neural network is combined. Based on the background of Shenzhen rainfall forecast, this paper puts forward the method of rainfall prediction for the city. The experimental results show this paper has achieved a high accuracy rate compared with the BP neural network, which provides a new way for urban rainfall forecasting.