Fuzzy logic in control and identification

    Fuzzy logic in control

This activity is aimed to investigate the application of the fuzzy logic paradigm for the control and indentification of dynamic system. In particular, fuzzy logic in control has been successfully used to capture heuristic control laws obtained from human experience or engineering practice in automated algorithm. These control laws are defined by means of linguistic rule, for example "if the pressure is high, then decrease the pump power". The heuristic approach in the controller design can be appealing for its simplicity, but formal design method can be mandatory in some cases. For this reason a great endeavor is carrying out by several researcher in defining formal design procedures.

    Fuzzy model identification

In literature a general approach to nonlinear structure modeling does not exist and then fuzzy models are interesting because they can approximate a large class of nonlinear functions. The maim problem consists in finding the parameters of the fuzzy model from data affected by noise. A well-established procedure, the Frisch Scheme, for linear identification in stochastic environment has been modified and exploited to be applied to fuzzy model identification. The extension of the implemented technique for the identification of multidimensional piece-wise linear causal models and piece-wise linear fuzzy models allows to identify nonlinear dynamic models.

Previous Page Simani Home Page Dipartimento di Ingegneria