A new differential evolution based on Gaussian sampling for forecasting urban water resources demand

In order to improve the performance of differential evolution (DE), this paper presents a new DE variant based on Gaussian sampling (NDEGS) to forecast urban water resources demand. In NDEGS, two strategies are employed. First, Gaussian sampling is used to replace the mutation operation. Second, a dynamic population method is employed to adjust the population size during the search process. In the simulation experiment, the water resources demand in Nanchang city of China is considered as a case study. Simulation results demonstrate that NDEGS can achieve promising prediction accuracy.