Short-term prediction of water demand provides basic guarantee for water supply system operation and management. In this study, an effective model for daily water demand forecasting is proposed. Firstly, principle component analysis(PCA) is utilized to simplify the complexity and reduce the correlation between influence variables, and the score values of selected principle components(PCs) turn into the irrelevant input data of fuzzy neural network(FNN), which models the prediction of water demand. Moreover, an improved Levenberg-Marquardt(ILM) algorithm is employed to optimize the parameters of FNN simultaneously, the problems of heavy computing burden and limited memory space can be solved. Most of all, a growing-pruning mechanism based on spiking integrate-and-fire(IF) model is applied to FNN in order to realize structural self-organization. Finally, contrast experiments are implemented to demonstrate that the spiking self-organizing fuzzy neural network(SSOFNN) has better prediction performance and capability to handle practical issues.