Fuzzy Inference Modeling Method Based On T-S Fuzzy System

Fuzzy inference modeling method can effectively describe complex systems, and this method is also widely used in system modeling. T-S fuzzy system can well construct the nonlinear system. The consequence of its inference rule is the linear or nonlinear function of the input information, so that each rule contains more information. As a result, one can use fewer rules to express more information of systems. Domestic and foreign scholars do lots of research on this subject. In this paper, we mainly study the dual-input single-output fuzzy system, and the simulation result of this method is compared with the fuzzy inference modeling methods based on other fuzzy systems. The main contents are as follows:1. Review the Mamdani fuzzy system, the fuzzy system based on fuzzy transformation and the T-S fuzzy system. Introduce the fuzzy inference modeling method and the fuzzy inference modeling method based on fuzzy transformation. Construct the input-output models and state-space models of time-invariant and time-variant by using these methods.2. Initially give the fuzzy inference modeling method based on the T-S fuzzy system. Construct the dual-input single-output T-S fuzzy system based on numerical differentiation method. The fuzzy inference modeling method based on the T-S fuzzy system is applied into the second-order freedom movement system modeling. Construct new input-output models (HX equation models) and new state-space models for the second-order systems respectively. Use these models to simulate Var der Pol equation and the second-order nonlinear time-varying differential equation y(t)-y(t)2+12ty=0.3. Compare the fuzzy inference modeling method based on the T-S fuzzy system with the fuzzy inference modeling method based on the Mamdani fuzzy system and the fuzzy transformation. The simulation part demonstrates the superiority of the proposed method.