Design optimization of double-acting hybrid magnetic thrust bearings with control integration using multi-objective evolutionary algorithms


Integration of the geometric and control designs in conjunction with the optimization is the current trend in mechatronic products. In the present work, an optimal design methodology of double-acting hybrid active magnetic thrust bearings has been proposed. Double-acting actuators and controller are optimized as a unified system. Conventionally, in control of rotors in the axial direction using double-acting magnetic bearings, two identical bearings are used. However, in the present design two different bearing geometries with different operating parameters have been considered. Minimization of the powerloss, the weight, the control input and dynamic performance indices and maximization of the load capacity have been considered as objectives. The design considers the 10 geometric, two electrical, and two control design parameters. The constraints are classified into three categories, namely the geometric, electrical, and control constraints. Real coded genetic algorithm has been implemented to carry out the constrained multi-objective optimization of the present problem. The convergence and Pareto-front spaces are studied by using different populations of sizes run for different generations. Some of the convergence criterions have been observed for actuator-controller systems. Designs which are nearest to the utopia point in Pareto-front fronts are compared. Air gaps, bias currents, and lengths of permanent magnets are observed to be consistently different for individual actuators of the double-acting bearing. Performance parameters of double-acting actuators and the controller of the magnetic bearing for different choices have been presented.