A multi-objective optimized control method for nonisolated dual three-phase permanent magnet synchronous motors with open-circuit fault

Abstract The zero-sequence current has long been a crucial challenge impeding the extensive application of non-isolated dual three-phase motors (NIDTP-PMSMs). To address this, a vector construction method based on model predictive control (MPC) is proposed, combined with a multi-vector, multi-space control strategy, enabling coordinated control of multiple subspaces current components and ensuring torque stability. By leveraging MPC’s ability to handle constraints and multi-variable systems, the impact of zero-sequence space currents on peak current and winding loss after single-phase open-circuit motor vector space decoupling (VSD) transformation is studied. A novel multi-objective optimization algorithm is then proposed to optimize the relationship between peak current and winding loss by adjusting the zero-sequence subspace current magnitude. This method can adjust the current amplitude in real time based on the peak current during motor operation, keeping the current below the maximum operating current and minimizing winding loss, reducing the risk of secondary accidents caused by overcurrent. Finally, simulations verify the effectiveness of the proposed strategy.