2026-05-30T01:07:34Zhttps://riuma.uma.es/rest/oai/request

oai:riuma.uma.es:10630/399822026-02-03T10:49:40Zcom_10630_2254col_10630_37953

Memory-based Model Predictive Control for Parameter Detuning in Multiphase Electric Machines González Prieto, Ángel González Prieto, Ignacio Dordevic, Obrad Aciego Gallardo, Juan José Montenegro, Jorge Durán-Martínez, Mario Javier Khan, Mohammad Control automático Motores eléctricos Ingeniería eléctrica Model predictive control (MPC) is a popular control technique to regulate multiphase electric drives (EDs). Despite the well-known advantages of MPC, it is sensitive to parameter detuning and lacks the capability to eliminate steady-state errors. The appearance of an offset between the reference and measured currents can significantly jeopardize the performance of the ED. This article suggests the use of a memory-based model predictive control (MB-MPC) that activates a compensation term when the parameter mismatch is detected. The suggested MB-MPC is universal for any multiphase machine if spatial harmonics are neglected since the proposed method does not consider any of the secondary x–y planes. Experimental results in two different rigs with six- and nine-phase induction motors prove this universality as well as its capability to eliminate current and speed offsets. 2025-09-19T09:08:25Z 2025-09-19T09:08:25Z 2025-09-19T09:08:25Z 2023-10-30 journal article A. Gonzalez-Prieto et al., "Memory-Based Model Predictive Control for Parameter Detuning in Multiphase Electric Machines," in IEEE Transactions on Power Electronics, vol. 39, no. 2, pp. 2546-2557, Feb. 2024, doi: 10.1109/TPEL.2023.3328427 https://hdl.handle.net/10630/39982 10.1109/TPEL.2023.3328427 eng PID2021-127131OB-I00 open access IEEE