The Construction And Control Application Of Novel Fuzzy Logic Systems Without Any Rules

The fuzzy logic system is an important research object in the research field of fuzzy control system theory, which has been widely applied in the control design of nonlinear uncertain systems. However, in the developing process of fuzzy control system theory, scholars of domestic and overseas paid more attention to conventional fuzzy logic systems based on IF-THEN fuzzy rules, especially Mamdani type and T-S type fuzzy logic systems, then used this kind of fuzzy logic systems to proceed adaptive control design for uncertain systems. In this kind of adaptive control design, on account of the structure of the controller and adaptive laws mainly depended on fuzzy rules, therefore, in order to improve the control precision and fuzzy approximation precision, a large number of fuzzy rules were usually needed to improve fuzzy approximation precision, which was easy to increase the online computation and result in large time delay, thus causing the system instability, that is, which was prone to generate "fuzzy rule explosion" problem. So how to solve this problem has become a hot research field of fuzzy control theory. It’s worth noting that among the designing schemes which were explored to solve this problem by scholars, there were few seek for corresponding solutions aiming at the fuzzy logic system structure itself.According to the above problem, this thesis mainly has done the following work:1. At the beginning, this thesis systematically introduces the fundamental knowledge of fuzzy logic system. Then sets about improving the structure of conventional fuzzy logic system, uses fuzzy sets to establish "language information base" which possess the characteristics of separating language variables to ameliorate the conventional IF-THEN fuzzy rules, focusing on the reconstruction of fuzzy relations, fuzzifiers and defuzzifiers. In the end, utilizes the method of unity partition function to integrate the fuzzy language knowledge to construct a novel fuzzy logic system without any rules.2. Apply the novel fuzzy logic system without any rules constructed above and Lyapunov stability control design method to design adaptive tracking controller for a class of nonlinear systems with uncertainties. And the effectiveness of the proposed method is verified through the inverted pendulum simulation examples.3. By introducing scalers and saturators with time-varying parameters, a kind of extended fuzzy logic system without any rules is constructed. Then based on this extended fuzzy logic system an adaptive stable controller is designed for a class of second-order uncertain nonlinear systems. Finally through the comparison of simulation results highlights the feasibility and superiority of this method.