Directive Adaptive Fuzzy Control For Nonlinear Uncertain Systems

In recent years, as the development of modern science and the explosion of productionscale, the complex industrial manufacture always shows the characters of uncertainty, strongnonlinear, multivariate, strong coupling and changeable industrial conditions, so the dynamicmodels of the systems cannot be described by accurate math models. Because fuzzy systemdo not need accurate math models and can use the experts’ knowledge effectively, thesystems have been used in a lot of real problems. By taking advantage of fuzzy logic systemand using human language information, three kinds of directive adaptive fuzzy controlstrategies based on uncertain nonlinear systems are proposed.In part one, a kind of direct adaptive fuzzy control strategy of nonlinear systems basedon generalized hyperbolic model is proposed. The kind of theory uses the generalizedhyperbolic model to replace the traditional triangle-shape grade of membership function orgauss membership function. Comparing with the traditional direct adaptive fuzzy control ofnonlinear systems, this theory can not only guarantee the system stability and the dynamicbehavior, but also can solve the problem of exponential growth of the fuzzy rules by takingadvantage of the special structure of generalized hyperbolic model whose number ofadjustable parameter is far less than the number of rules. So the control method can reducethe computing and increase the fractional application. The simulation results prove theadvantage.In part two, a kind of directive adaptive fuzzy control strategy based on Backsteppingdesign is proposed, which can combine Backstepping design and fuzzy logic system.Comparing with existing adaptive fuzzy control strategy based on Backstepping design, thisstrategy only uses a fuzzy logic system to approximate the uncertain function, so this theorysave the computing. In addition, this strategy can improve the robust, by using Lyapunovmethod. It can not only guarantee that all the signals are uniform bound, but also it can makethe estimate value of tracking error in a small zero neighborhood. Simulation results provethe effect of strategy.In part three, a kind of robust adaptive fuzzy control strategy is proposed, which designsa rubost compensator. The strategy can elimate the influence of approximate errors andexternal disturbance. And at the same time the algorithm gerantee the stability of the system.The simulation results prove the unility.