Control of Flexible Rotor by Using Electromagnetic Actuators Optimization of the Fuzzy Controller Gains

The aim is to set an optimization procedure for the determination of a fuzzy based controller gains. The objective function is defined as a combination of the maximum current, the displacements measured and the settling time. Two fuzzy based controllers were developed, the first has displacement and velocity as inputs (a fuzzy Proportional Derivative) and the second utilizes inputs expressed in polar coordinates. Its originality is that it manages two significant physical quantities, namely tangential and radial velocities, associated with steady state and transient behaviours respectively. The gains of the fuzzy PD controllers were determined by using optimization procedures that takes into account the mechanical behavior and the energy consumption. The gains (for the three controllers) were first determined by using numerical simulations, and then implemented for experimental testing. An additional tuning (in-situ) was necessary for the PD controller. The performances of the two fuzzy based controllers were compared to a PD controller experimentally from mechanical and energy consumption points of view. The three controllers were efficient with a better behavior for the fuzzy based controllers for run up conditions. The optimization procedures are efficient and easy to implement in the case of PD fuzzy controllers, and must be enhanced to consider rules with different output, as in the case of polar fuzzy controller.