Optimization-Based Robust Control for Momentum Wheel Systems Subject to Large-Scale Power Supply Fluctuations

Voltage fluctuations in unregulated microsatellite buses introduce strong coupled nonlinearities in momentum wheel systems, which consist of power supply (PS) and permanent magnet synchronous motor (PMSM) subsystems. Voltage fluctuation-induced coupling can dramatically affect the performance of momentum wheel systems. To address this specific challenge, this article proposed a systematically dual-loop controller based on feedback linearization, mixing H₂/H_∞ and linear quadratic regulator (LQR) control strategies. For the inner loop control, the affine nonlinear PS-PMSM model is linearized using differential geometric techniques, and mixed H₂/H_∞ control is applied to this linearized coupled model to guarantee robust stability and dynamic response characteristics. Regarding the outer loop control, we propose an LQR framework integrated with Gaussian distribution particle swarm optimization (GDPSO) algorithm for parameter tuning. Simulation results show that the proposed controller has a 64% reduction in speed step response time, 51% mitigation of DC-bus voltage fluctuations, and the proposed GDPSO achieves 50% faster convergence compared to conventional particle swarm optimization methods. Hardware experiments also validate superior performance, with a total harmonic distortion of only 7.96%, offering an effective solution for high-precision control of multivariable nonlinear electric servo systems.