关键词: 组合预测, 模糊系统理论, Takagi－Sugeno规则

Abstract: In this paper, a new nonlinear combination forecasting method based on fuzzy Takagi-Sugeno model system is presented to overcome the limitation in linear combination forecasting. Furthermore, the corresponding gradient descent learning algorithm is put forward to identify the parameter of the fuzzy model and partitions of fuzzy subsets. Theoretical analysis and forecasting examples all show that the new techniques has reinforcement learning properties and universalized Capabilities. With respect to combined modeling and forecasting of non-stationary time Series in nonlinear systems, which have some uncertainties, the method are feasible and effective.