25-09-2010, 01:28 PM
Torque Ripple Minimization Of Switched Reluctance Machine
Using Genetic Algorithm
A major problem of Switched Reluctance Motors (SRM) is the Torque ripple, which causes undesirable acoustic noise and vibration. saliency of the stator and rotor is the main cause for this. This article describes an algorithm for reducing the torque dip on the basis of Genetic Algorithm (GA), thereby reducing the associated torque ripple.
Introduction
because of its simple mechanical structure and development of power electronics, the switched reluctance machines are used many high speed applications. nonlinear property of the inductance according to the current and rotor position is the major cause for producing noise. servo or servo-type systems are intolerant to torque ripple and hence thay must be reduced.reduction in the average torque may result when electronic control is used.
Torque ripple Minimization
reduction in torque ripple is possible by splitting the stator poles
and skewing the rotor poles. The phase overlap regions have the majority of the torque ripple . In this region, the torque producing
responsibility is commutated from one phase to another. The definition of torque ripple is :-
"difference between the maximum and minimum instantaneous torque expressed as a percentage of the average torque during steady state operation"
Genetic algorithm has computational procedures that are just like the evolution. The advantages of using it are:
-They are adaptive, and learn from experience.
-They have intrinsic parallelism
-They are efficient for complex problems.
To solve the problem of torque dip, Flux linkage is maximized with inductance overlap ratio of two adjacent phases and thus improve the performance of a 3phase SRM system.
Here the The design variables are Stator pole arc and
Rotor pole arc. Best values are found for the Stator and rotor pole arcs. Application of the genetic method provides the best result in the least time.
report:
[attachment=4375]