TIME-VARYING SYSTEM MODELING METHOD BASED ON FUZZY INFERENCE
基于Fuzzy推理的时变系统建模

A kind of modeling method for general time-varying systems, which uses modeling method based on fuzzy inference (MMFI), is proposed here. To get model of a time-varying system, firstly the time dimension divided into some periods, then in every short time interval, time-invariant model is used to approach the original time-varying system, in the end a time-varying nonlinear differential equation with variable coefficients is obtained. The processes of how to obtain the input-output type model and the state space type model are discussed in detail. In addition, a theory is given to ensure the approximation performance of HX equations. The simulation experiments show that the mathematical models formed by this method can approximate the general time-varying nonlinear systems with high accuracy.