Search

In the field of speed sensorless control of permanent magnet synchronous motors, it is difficult to accurately estimate the speed via a single algorithm for full speed range because there are many non-ideal factors when the algorithms are realized. In this paper, high frequency pulse signal injection method is applied in the low speed region, and dual EKF is used in the medium and high speed region. The switching of these two algorithms is achieved by the weighted averaging rule. This hybrid estimation strategy avoids detecting the back-electromotive force which is hard to be measured in low-speed region, and also avoids the adverse effects of the filter delay in the mid-high speed region. Therefore, it enables the sensorless control effective over the full speed range. Moreover, the dual EKF algorithm can estimate the motor parameters with a relatively small amount of calculation. Validity and effectiveness of the proposed scheme is shown in the simulation results.

In the daul PWM converter system of permanent magnet synchronous generator, the passivity can achieve global asymptotic stability and improve the robustness to external disturbances of the system. However, the power ripple on the grid-side is still large. Aiming at this problem, an model predictive control with fixed switching frequency is proposed in this paper. The mathematical model of the grid-side converter is established and analyzed, then the cost functions of the model predictive control is selected based on the active and the reactive power error. Furthermore, according to the analysis and the derivation of the cost function, SVPWM technique is applied to achieve the fixed switching frequency, so as to reduce the switching losses. The simulation and the experimental results show that the proposed model predictive control scheme can depress the power ripple on the grid-side effectively.