Real-time optimal torque control of permanent magnet synchronous motors

Abstract

In the dissertation, a new method is proposed for real-time optimal torque control of permanent magnet synchronous motors (PMSMs). Numerical optimization techniques are introduced to solve the optimal torque control problem that is a constrained nonlinear optimization, greatly reducing computation time compared to the analytic solution and a commercial numerical solver. Unlike existing numerical solvers to the optimal torque control problem, convergence of the proposed numerical solver is ensured for all operating regions of PMSM. A real-time parameter estimation algorithm is also presented to overcome the PMSM parameter dependency of the proposed numerical solver. The rank deficiency problem of the PMSM state equations is solved by adding the newly derived stator current ripple model, which enables real-time estimation of all PMSM parameters. The performance and practicality of the proposed method will be verified by experimental studies with a 7.5-kW interior PMSM.