The ill-convergence of multiparameter estimation due to the rank-deficient state equations of permanent-magnet synchronous machines (PMSMs) is investigated. It is verified that the PMSM model for multiparameter estimation under i _d = 0 control is rank deficient for simultaneously estimating winding resistance, rotor flux linkage, and winding inductance and cannot ensure them to converge to the correct parameter values. A new method is proposed based on injecting a short pulse of negative i _d current and simultaneously solving two sets of simplified PMSM state equations corresponding to i _d = 0 and i _d ≠ 0 by using an Adaline neural network. The convergence of solutions is ensured, while the minimum | i _d | is determined from the error analysis for nonsalient-pole PMSMs. The proposed method does not need the nominal value of any parameter and only needs to sample the winding terminal currents and voltages, and the rotor speed for simultaneously estimating the dq -axis inductances, the winding resistance, and the rotor flux linkage in nonsalient-pole PMSMs. Compared with existing methods, the proposed method can eliminate the estimation error caused by the variation of rotor flux linkage and inductance as a result of state change due to the injected d -axis current in the surface-mounted PMSM. The method is verified by experiments, and the results show that the proposed method has negligible influence on output torque and rotor speed and has good performance in tracking the variation of PMSM parameters due to temperature variation.

This paper proposes a method for online estimating the rotor flux linkage and voltage-source inverter (VSI) nonlinearity of permanent magnet synchronous machine (PMSM) drives. Thermocouples are employed for measuring the temperature variation of the stator winding in order to obtain the actual value of stator winding resistance. An Adaline estimator is used for online estimation of distorted voltage Vdead due to VSI nonlinearity. Both are subsequently used for the estimation of the rotor flux linkage. The proposed method is experimentally validated on a PMSM drive system and shows good performance in tracking the variation of the rotor flux linkage and compensating the VSI nonlinearity.

Online parameter estimation of permanent magnet synchronous machines is critical for improving their control performance and operational reliability. This paper provides an overview of the recent achievements of online parameter estimation of PMSMs with examples. The critical issues in parameter estimation are firstly analysed, especially the rank-deficient issue and inverter nonlinearities. Then, the state-of-the-art online parameter estimation modelling techniques are reviewed and assessed. Finally, some typical applications and examples are outlined, e.g. estimation of mechanical parameters, improvement of sensored and sensorless control performance, thermal condition monitoring, and fault diagnosis, together with future research trends.

A global parameter estimation method for a permanent magnet synchronous machines (PMSM) drive system is proposed, where the electrical parameters, mechanical parameters, and voltage-source-inverter (VSI) nonlinearity are regarded as a whole and parameter estimation is formulated as a single parameter optimization model. A dynamic learning estimator is proposed for tracking the electrical parameters, mechanical parameters, and VSI of PMSM drive by using dynamic self-learning particle swarm optimization (DSLPSO). In DSLPSO, a novel movement modification equation with dynamic exemplar learning strategy is designed to ensure its diversity and achieve a reasonable tradeoff between the exploitation and exploration during the search process. Moreover, a nonlinear multiscale based interactive learning operator is introduced for accelerating the convergence speed of the Pbest particles; meanwhile a dynamic opposition-based learning strategy is designed to facilitate the gBest particle to explore a potentially better region. The proposed algorithm is applied to parameter estimation for a PMSM drive system. The results show that the proposed method has better performance in tracking the variation of electrical parameters, and estimating the immeasurable mechanical parameters and the VSI disturbance voltage simultaneously.

Winding resistance and rotor flux linkage are important to controller design and condition monitoring of a surface-mounted permanent-magnet synchronous machine (PMSM) system. In this paper, an online method for simultaneously estimating the winding resistance and rotor flux linkage of a PMSM is proposed, which is suitable for application under constant load torque. It is based on a proposed full-rank reference/variable model. Under constant load torque, a short pulse of id ≠ 0 is transiently injected into the d -axis current, and two sets of machine rotor speeds, currents, and voltages corresponding to id = 0 and id ≠ 0 are then measured for estimation. Since the torque is kept almost constant during the transient injection, owing to the moment of system inertia and negligible reluctance torque, the variation of rotor flux linkage due to injected id ≠ 0 can be taken into account by using the equation of constant torque without measuring the load torque and is then associated with the two sets of machine equations for simultaneously estimating the winding resistance and rotor flux linkage. Furthermore, the proposed method does not need the values of the dq -axis inductances, while the influence from the nonideal voltage measurement, which will cause an ill-conditioned problem in the estimation, has been taken into account and solved by error analysis. This method is finally verified on two prototype PMSMs and shows good performance.

This paper proposes a novel scheme for the identification of the whole flux linkage map of permanent magnet synchronous machines (PMSMs), by which the map of dq-axis flux linkages at different load or saturation conditions can be identified by the minimization of a proposed estimation model. The proposed method works on a conventional three-phase inverter based vector control system and the immune clonal based quantum genetic algorithm (ICQGA) is employed for the global searching of minimal point. Besides, it is also noteworthy that the influence of uncertain inverter nonlinearity and circuit resistance are cancelled during the modeling process. The proposed method is subsequently tested on two PMSMs and shows quite good performance compared with the finite element prediction results.