Model and application of annual river runoff prediction based on complementary set empirical mode decomposition combined with particle swarm optimization adaptive neuro-fuzzy system

Abstract Runoff is affected by natural and nonnatural factors in the process of formation, runoff series generally nonstationary time series. How to improve accuracy prediction has always been a difficult problem for hydrologists. The key solve this reduce complexity model. Based on aforementioned ideas, article uses complementary set empirical mode decomposition decompose into multiple intrinsic components that retain time–frequency information, thus reducing particle swarm optimization (PSO) adaptive neuro-fuzzy system used predict each component prediction. After that, trained model are reconstructed original example shows absolute relative error forecasting constructed 0.039, determination coefficient 0.973. This can be applied annual forecasting. Comparing results with algorithm-ANFIS ANFIS model, algorithm-PSO-ANFIS obvious advantages.