Urban water demand prediction method

The invention belongs to the technical field of urban water demand prediction, particularly discloses an urban water demand prediction method, and solves the problems that an existing water demand prediction method does not need more data, but is difficult to fully consider in prediction due to the fact that an actual water consumption system is very complex and factors influencing water consumption demands are numerous. In order to solve the problems of large deviation between the prediction result of the method and the actual water consumption due to the fact that the time series of urban water demand data is short, main influence factors of water demand are identified by using a principal component analysis method, a single prediction method is optimized based on a grey correlation theory, and for the limitation of the single prediction method, the prediction result is optimized. The optimal prediction method is weighted by adopting the variable weight combination prediction model, the prediction precision