Design and simulation of proportional-derivative controller of a servo systems based on fuzzy logic and hybrid fuzzy neural

Abstract EN

In the past few years, fuzzy-rule-based modeling has become an active research field because of its good merits in solving complex nonlinear system identification and control problems.

A servo system (SS) is a class of a nonlinear position system that needs to be positioned accurately and fastly on a commanded position.

The strategy followed in this paper in designing digital controller for such system ; first building a neuro-model that represents the open loop servo system.

This is accomplished by sufficiently collecting input-output data and used it off-line to build the neural network that will represent the plant for the second design stage.

Second design fuzzy controller through NN simulation to reach the required closed–loop behavior.

The design technique is based on the adjustment of the scale factors, rule base and membership functions of the controller; it and was accomplished by fine tuning and heuristic corrections linked to the knowledge of the process to be controlled.

For the specified plant, there are certain parameters, which produce a well-controlled response.