To read this content please select one of the options below:

(excl. tax) 30 days to view and download

Multi-objective optimization of bias current coefficient based on NSGA- III for active magnetic bearing with redundant electromagnetic actuators

Shuai Deng, Xin Cheng, Huachun Wu, Yefa Hu

Engineering Computations

ISSN: 0264-4401

Article publication date: 11 July 2024

Issue publication date: 2 August 2024

48

Abstract

Purpose

The multi-objective optimization configuration strategy is proposed due to the configuration of EMAs in fault-tolerant control of active magnetic bearing with redundant electromagnetic actuators involving high-dimensional, nonlinear, conflicting goals.

Design/methodology/approach

A multi-objective optimization model for bias current coefficients is established based on the nonlinear model of active magnetic bearings with redundant electromagnetic actuators. Based on the non-dominated sorting genetic algorithm III, a numerical method is used to obtain feasible and non-inferior sets for the bias current coefficient.

Findings

(1) The conflicting relationship among the three optimization objectives was analyzed for various failure modes of EAMs. (2) For different EMAs' failure modes, the multi-objective optimization configuration strategy can simultaneously achieve the optimal or sub-optimal effective EMF, flux margins, and stability of EMF. Moreover, the characteristics of the optimal Pareto front are consistent with the physical properties of the AMB. (3) Compared with the feasible configuration of C0, the non-inferior configurations can significantly improve the performance of AMB, and the advantages of the multi-objective optimization configuration strategy become more prominent as the asymmetry of the residual supporting structure intensifies.

Originality/value

i) Considering the variation of the rotor displacement during the support reconstruction, a decision-making model that can accurately characterize the dynamic performance of AMB is presented. (ii) The interaction law between AMB and rotor under different failure modes of EMAs is analyzed, and the configuration principles for redundant EMAs are proposed. (iii) Based on the dynamic characteristics of AMB during the support reconstruction, effective EMF, energy consumption, and the Pearson correlation coefficient between the desired EMFs and the decoupled control currents are used as objective functions. iv. The NSGA-III is combined with the decision-making model to address the multi-objective optimization configuration problem of C0.

Keywords

Acknowledgements

This work is partly supported by the National Natural Science Foundation of China 52275065 and partly by the National Key R&D Program of China with Grant No. 2018YFB2000103.

Citation

Deng, S., Cheng, X., Wu, H. and Hu, Y. (2024), "Multi-objective optimization of bias current coefficient based on NSGA- III for active magnetic bearing with redundant electromagnetic actuators", Engineering Computations, Vol. 41 No. 6, pp. 1549-1571. https://doi.org/10.1108/EC-03-2023-0127

Publisher

:

Emerald Publishing Limited

Copyright © 2024, Emerald Publishing Limited

Related articles