PI parameterization using adaptive fuzzy modulation (AFM) for networked control systems. II. Full adaptation

For pt.I see ibid., p.3152-7 (2002). The partial adaptive fuzzy modulation (AFM) applied to IP networked PI control systems were given in Part I. In Part II of this paper, we propose a full adaptive fuzzy modulation control scheme and compared it with two other tuning methods to compensate for the IP network-induced time delays in IT networked PI control systems. In the first tuning method, we adaptively tune the PI controller parameters directly. In the second tuning method, we use partial adaptive fuzzy modulation (AFM), i.e., tune the consequent parameters in the fuzzy rules only with a fixed membership function parameters in the antecedent part and fixed PI controller parameters to their nominal values. In the full AFM, not only the consequent parameters to be tuned adaptively, but also the membership functions parameters in the antecedent part are tuned adaptively. Numerical simulation of a network-based controlled DC motor is used to illustrate the effectiveness of the proposed full AFM control scheme over other controllers.