Development of Small and High-Efficiency Self-Bearing Motor Using Self-Sensing Technique for Artificial Heart

本研究为IPM型自承式电机提出了一种新型的自感知控制，该电机具有高效率和高设计可能性的永磁体结构。由于主磁通方向的磁阻比正交磁通方向的磁阻大，IPM电机具有磁凸极特性。因此，当转子旋转时，电机绕组的电感会出现变化问题，不能利用这种变化进行自感知。为了克服这一问题，提出了d-q等值结构，并证明了电感变化可以忽略不计。然后推导出基于磁等效电路的近似方程，解释了差动输出的非线性特性。结果表明，旋转电流和悬浮电流对传感器的传感性能有不利影响。分别对旋转电流和悬浮电流进行了非线性补偿。该补偿实现了较好的位移估计。但是，即使个人补偿是有效的，两种补偿结果的叠加也不能有效地发挥作用。最后，针对IPM自承式电机的稳定悬浮和旋转问题，提出了一种基于二元多项式近似系数的线性化方法。

This research proposes new self-sensing control for the IPM type self-bearing motor possessing high efficiency and high design possibility of permanent magnet configuration. IPM motor has a magnetic salient pole property because of the magnetic reluctance of the main flux direction is bigger than the quadrature flux direction. Therefore, when the rotor rotates, the inductance variation problem of the motoring winding occurs and self-sensing could not use by the variation. To overcome this problem, d-q equivalent structure was proposed and confirmed the inductance variation was negligible. Then the approximate equation based on the magnetic equivalent circuit is derived to explain the nonlinear characteristics of the differential output. The result shows the rotation and levitation currents have an adverse effect of sensing property. The nonlinear compensation is introduced for rotation and levitation current individually. The compensation realized good displacement estimation. But, even the individual compensation is effective, the superposition of the two compensation results does not work effectively. Finally, linearization method using the approximate coefficients of polynomial equation with two unknowns is proposed for stable levitation and rotation of the IPM self-bearing motor.